Key findings of the study include:

Transcription

1 C I T Y O F C A M B R I D G E Community Development Department IRAM FAROOQ Assistant City Manager for Community Development MEMORANDUM To: Richard Rossi, City Manager From: Iram Farooq, Assistant City Manager for Community Development Date: April 6, 2016 Re: Inclusionary Housing Study We are pleased to transmit the Cambridge Inclusionary Housing Study recently completed by David Paul Rosen & Associates. The attached study provides in depth analyses and recommendations for changes to the City s Inclusionary Housing Ordinance by: reviewing the current inclusionary housing provisions of the Zoning Ordinance and comparing these standards to inclusionary housing programs in other localities; analyzing changes in the socioeconomic profile of the city and housing affordability in the years since the adoption of the inclusionary housing provisions in 1998; modeling the economic impact of the current program on typical developments seen in the City and estimating the potential impact of increased set-asides of affordable housing; and, outlining policy options and recommendations for consideration as changes to the inclusionary housing provisions are discussed. Key findings of the study include: 344 Broadway Cambridge, MA Voice: Fax: TTY: increases in market rents and sales prices have outstripped increases in income in recent years; affordable housing created under the inclusionary housing provisions has become an increasingly critical source of new affordable housing as other mechanisms to expand the affordable stock have become more challenging; strengthening the inclusionary housing provisions is necessary to maintain an adequate stock of affordable housing and preserve the socioeconomic diversity of the city; inclusionary housing provisions may be increased to a certain extent without severely compromising the production of new market-rate housing.

2 Background The current inclusionary housing provisions were adopted in 1998 after two City-commissioned studies were completed. In one study completed in 1997, Peter Werwath and Associates recommended a framework for inclusionary housing provisions which became the basis for the current inclusionary housing provisions in the Zoning Ordinance. Then, a rationale study completed by Stockard, Engler & Brigham in 1998 recommended that for every 10 new market-rate housing units, 1.5 additional affordable units should be created to preserve the economic diversity of the community. The inclusionary housing provisions apply to housing developments of 10 or more units, which include units that are affordable to households earning less than 80% of Area Median Income (AMI). The current provisions set a 15% affordable housing ratio. Most developments are eligible for an allowed 30% density bonus, which typically results in an effective affordable housing ratio of approximately 11.5% of units in many new market buildings. Developments that are not eligible for the density bonus provide a full 15% of total units as affordable. The density bonus is an important and intentional provision for which there is legal support. The application of the density bonus, however, has on occasion caused confusion about the expected number of affordable units in new buildings. There are currently 891 affordable units completed or now being developed through the inclusionary housing and similar zoning provisions. Inclusionary units have grown to become a significant component of the City s affordable housing stock and have been a critical means to create new units in recent years. The Cambridge inclusionary housing provisions were cutting edge when adopted, and continue to be cited as a successful model for other communities. While the Cambridge program has been successful at creating new permanently affordable units, we wanted to examine the program in the context of current housing needs. In 2014, CDD commissioned David Paul Rosen & Associates (DRA) to conduct a comprehensive new inclusionary housing study with the following objectives: to analyze the impact of new market rate residential development on housing affordability, housing supply, housing needs, demographic trends and socioeconomic diversity in Cambridge; to recommend changes to the inclusionary housing provisions contained in Article of the Cambridge Zoning

3 Ordinance; and to provide advice on current needs and best practices from inclusionary housing programs in other communities. Founded in 1980, DRA provides affordable housing advisory services to public agencies, private lenders and investors, and residential developers. DRA has completed dozens of similar studies analyzing the economic impact of inclusionary housing provisions, nexus fees, housing mitigation measures, and regulatory reform on development in communities across the country, and has helped more than forty communities research and develop inclusionary housing programs and other affordable housing mitigation measures. Comparison to Other Communities DRA reviewed the Cambridge inclusionary housing provisions and compared these to similar provisions in other communities. DRA s comparison included communities selected by CDD staff along with communities identified by DRA as having model programs or policies and/or similar needs which would provide a worthy comparison. Communities included in this comparison included: Boston; Boulder, Colorado; Burlington, Vermont; San Francisco; Santa Monica, California; Somerville; and Washington, DC. Overall, Cambridge s provisions compared favorably with those in other communities. Cambridge is unusual in that affordable units are built on-site in all cases, and are comparable to market units in terms of location, quality and size. While the typical effective set-aside ratio of 11.5% in Cambridge is middle of the pack compared to surveyed communities, DRA found that communities with higher set-aside ratios typically targeted higher income households for affordable units, and those with lower set-aside ratios targeted lower-income households. Unlike Cambridge, all but one other surveyed program had higher income limits for ownership than for rental housing. Density bonuses were offered in half of the surveyed programs. DRA also found that Cambridge s provisions are notable for their consistency and predictability, and for their success in creating mixed-income communities where high-quality affordable units are indistinguishable from market units. Socioeconomic Diversity Analysis DRA conducted a socioeconomic diversity analysis which analyzed how housing affordability and socioeconomic diversity in the city has changed in recent years by examining trends in market housing costs, affordable housing, household income distribution, and the affordable housing stock. DRA found

4 that increases in market-rate rents and sales prices have outpaced increases in income, resulting in a reduction of overall housing affordability. Household income needed to afford market-rate rental and ownership housing has increased, making it more and more unlikely that a low or moderate-income households can find an affordable home on the market. Cost burdening (i.e., paying more than 30% of a household s gross income on housing) remains a significant issue for households earning under 80% AMI, and has become an increasing issue even for households earning up to 100% AMI. DRA found that the household income characteristics of the city have changed significantly since The biggest changes were reductions in the proportion of moderate-income households (i.e. incomes between 50% and 80% AMI) which declined from 13% of the community in 2000 to 8% in 2011, and lower-middle-income households (i.e. incomes between 80% and 100% AMI) which declined from 14% in 2000 to 8% in These changes occurred for both renters and owners but were more dramatic for renters. These changes were offset by an increase in the ratio of households earning over 120% AMI (increasing from 35% of the community in 2000 to 47% in 2011). The proportion of households with incomes under 50% AMI showed little change in this comparison. DRA found that despite market pressures, the City has been successful in maintaining the overall ratio of affordable housing in the city at roughly 15% of the total housing stock. This has been accomplished through various housing strategies, including the current inclusionary provisions and other City-funded affordable housing. DRA noted, however, that an increasing proportion of the City s stock of affordable units are targeted to households earning less than 60% of AMI, making the inclusionary housing program one of the few mechanisms that caters to households earning up to 80% of AMI. Economic Feasibility Analysis DRA modeled alternative inclusionary housing set-aside standards for seven development prototypes based on typical development models. Development prototypes included large-scale rental buildings (high-rise and low-rise), moderate-scale rental and homeownership buildings, and smaller-scale rental and homeownership buildings. The economic modeling was intended to estimate the financial impact of differing set-aside standards on different types of developments to help inform discussions about increases in inclusionary setaside ratios. No specific development project was modeled. The results are

5 intended to be a guide for discussing the potential impact of regulatory changes on the feasibility of market residential development. DRA consulted with local developers and other experts to get information on land, construction and other development costs typical for each development type. Financial modeling was done using both a return on equity (ROE) and residual land value (RLV) approach for each scenario with changes in affordability set-asides. Affordability set-asides varied from the current provisions (typically 11.5%, after application of the density bonus) up to 25% affordable with varying affordability targets. For purposes of this modeling, DRA assumed that the current 30% inclusionary density bonus was available in each scenario and that no additional zoning offsets would be available (e.g. additional density or height or relaxed dimensional requirements). Cambridge s current housing market is exceptionally strong, as demonstrated by high demand and historically low cap rates. Cap rates are a measure of observed property sales which can be used to estimate the value of a property based on income it produces. Low cap rates in Cambridge indicate that residential property in Cambridge is a very attractive investment and that there is very strong demand for buildable land. Given the strength of the current market, DRA considered two market scenarios for each prototype, using both a cap rate that is typical of the current market and a slightly higher cap rate, in acknowledgement that cap rates seen in the current market are lower than long-term averages and that developments which are feasible today might not be as feasible in a different market climate. DRA s analysis found that in the current market, increasing the affordability set-aside up to 20% of the total units in the prototypical developments modeled would have a financial impact, but would not necessarily render developments economically problematic. Reflecting standard industry practices, feasibility was set at a threshold of 8% ROE or a RLV at or above market land costs. Policy Options and Recommendations Based on changes described in its socioeconomic diversity analysis, the economic feasibility analysis of prototypical developments with varying affordability requirements, and its review of inclusionary programs in other communities and national best practices in inclusionary housing, DRA s recommendations for Cambridge include:

6 Increasing the set-aside ratio for affordable units up to a net of 20% of total units built in a residential project; For affordable inclusionary rental housing, either maintaining the current income eligibility limit of 80% of AMI for the recommended inclusionary set-aside (i.e. up to 20% under 80% AMI), or making up to 15% of units affordable to households earning less than 80% AMI and up to 5% of units affordable to households earning less than 100% AMI; For affordable inclusionary ownership housing, increasing the eligibility limit to 100% of AMI and increasing affordability targets above 65% of AMI; Maintaining the 30% allowed increase in density for an inclusionary housing project; Considering a higher set-aside ratio in unique cases through zoning ordinance changes where warranted; Considering specifying that a portion of affordable units created must be 3-bedroom units, and/or that inclusionary housing set-aside ratios can be applied on a per-bedroom or per-square-foot basis. Disallowing the provision of affordable studio units, or creating a pricing structure in which studio units can be offered at a lower cost than one-bedroom units; Considering lowering the unit and/or square footage threshold for triggering the inclusionary housing provisions; For projects which result in at least one inclusionary unit and a fractional unit, permitting an in-lieu-of contribution for the fractional affordable unit rather than rounding the number of affordable units up or down; Allowing the option to select inclusionary units that are less premium in a development in exchange for increasing the total number of affordable units; Considering the impacts of the changes to the inclusionary housing provisions on the competitiveness of residential versus commercial development and land uses; Applying changes to the inclusionary housing provisions only to new developments. Other Considerations While the report contains a number of DRA recommendations which provide options for City discussions, we would also note a few other considerations to be mindful of when considering changes to the inclusionary housing provisions.

7 The DRA report focused on housing affordability and changes in the housing market and community since the current inclusionary provisions were adopted. There are other policy, legal and practical considerations to be aware of as we discuss recommendations and consider changes to the inclusionary provisions. It is important to consider not just the impact changes might have on overall housing production, but also other City goals such as balancing residential and commercial development, achieving excellence in urban design, reducing adverse transportation impacts, setting high standards for environmental sustainability, and engaging community members in a meaningful review of development proposals. Finally, given the depth of this study, the myriad factors to consider as changes are discussed, the impact of inclusionary housing provisions on our ability to meet affordable and other housing goals and on the feasibility of new housing development, and other public policy goals in the city, we recommend that the City reassess the inclusionary housing provisions periodically. We recommend that the City undertake an update in five years to assess the impact of any changes made now, and to determine whether and how the inclusionary housing provisions are addressing the housing needs and priorities of the city at that time. Next Steps Increasing the effective ratio of inclusionary housing would be a significant step forward in addressing the continuing need for affordable housing in the community. A thoughtful increase which does not significantly alter the development landscape and continues to support the development of housing, would greatly assist in offsetting changes in socioeconomic diversity of the community from higher income households moving into new market-rate housing in the city. We look forward to discussing this study and its recommendations with the City Council and community. In coordination with the City Council, we will reach out to and be available to meet with residents, developers, and others interested in the study. We can gather and prepare feedback on the report if that is helpful to the City Council. We will also review and discuss this study with the Affordable Housing Trust and ask that they be available to assist the City Council with this review.

8 As the City Council s discussion moves forward and consensus is reached on a desired set of policy changes, we will work to draft a zoning amendment that would implement the desired changes to the City s inclusionary housing provisions.

9 Final Report Cambridge Inclusionary Housing Study March 27, 2016

10 i SUBMITTED TO: City of Cambridge Community Development Department 344 Broadway Cambridge, MA SUBMITTED BY: David Paul Rosen & Associates 1330 Broadway, Suite 937 Oakland, CA Fax Hendrix Street Irvine, CA Fax Final Report i

11 Table of Contents Executive Summary and Introduction... 1 Summary of Key Findings... 2 Contents of Report... 4 Detailed Summary of Findings... 5 Findings of Socioeconomic Diversity Analysis... 5 Findings of Economic Feasibility Analysis... 7 Existing Inclusionary Zoning Program Background and History Inclusionary Provisions of the Zoning Ordinance Inclusionary Housing Programs in Selected Communities Socioeconomic Diversity Analysis Affordable Housing Income Limits Market-Rate Housing Rents and Sales Prices Housing Affordability Housing Cost-Burden Household Income Distribution Impact of Market Rate Housing on Income Diversity Affordable Housing Stock Household Tenure and Size Bedroom Count Distribution and Family Sized Units Economic Feasibility Analysis Methodology and Definition of Key Terms Development Prototypes Inclusionary Housing Scenarios Estimated Prototype Development Costs Land Acquisition Costs Hard Construction Costs Soft (Indirect) Development Costs Total Development Costs Estimated Market Rents and Sales Prices Final Report ii

12 Apartment Rents, Vacancy Rates and Operating Costs Condominium Sales Prices Affordable Rents and Sales Prices Inclusionary Housing Scenarios and Calculations Return on Equity and Land Residual Analysis Results Policy Options and Recommendations List of Charts 1. Income Distribution, City of Cambridge, 2000 and Final Report iii

13 List of Tables Table Number Table Title Table Subheading Page Number Table 1 Summary of Development Prototypes City of Cambridge 6 Inclusionary Housing Study Table 2 Summary of Return on Equity and Land Residual Residential Development Low Cap Rate 10 Analysis Results Prototypes Assumption Table 3 Summary of Return on Equity and Land Residual Residential Development High Cap Rate 11 Analysis Results Prototypes Assumption Table 4 Inclusionary Housing Set-Aside, Density Bonus and Threshold Size Provisions Survey of Selected Communities Table 5 Affordable Housing Income Limits by Household Size City of Cambridge Inclusionary Housing Study Table 6 Median Market-Rate Asking Rents City of Cambridge Rent 1997 to Survey Table 7 Annual Compound Growth Rate in Median Market- City of Cambridge Rent 2000 to Rate Asking Rents Survey Table 8 Median Sales Prices for Single-Family Homes and City of Cambridge 2007 to Condos Assessing Data Table 9 Compound Annual Growth Rate in Median Sales City of Cambridge 2008 to Prices for Single-Family Homes and Condos Assessing Data Table 10 Trends in Area Median Income and Affordable Rents City of Cambridge 1997 to Table 11 Affordability of Median Asking Rents for a Four-Person City of Cambridge Rent Household Survey Table 12 Annual Gross Income Required to Afford Median City of Cambridge 1997 to Asking Rents for One-, Two-, and Three-Bedroom Units Table 13 Annual Gross Income Required to Afford Media- City of Cambridge 2007 to Priced Homes and Condos Table 14 Percentage of Cost-Burdened Households by Income City of Cambridge 2000 to Level Table 15 Distribution of Households by Income Level City of Cambridge 2000 to Table 16 Distribution of Households by Household Size and City of Cambridge 2000 to Income Level Table 17 Affordable Housing Inventory City of Cambridge 2008 to Table 18 Proportion of Housing Stock with Long-Term City of Cambridge 1997 to Restrictions Affordable to Low and Moderate Income Residents Table 19 Household Size Distribution City of Cambridge 2000 to Table 20a Distribution of Housing Units by Bedroom Count City of Cambridge 2000 to Table 20b Distributionof Housing Units by Bedroom Count; City of Cambridge September, Projects Under Inclusionary Housing Covenants Table 21 Development Prototypes 47 Table 22 Inclusionary Housing Scenario Set-Aside and Income Cambridge Inclusionary Targeting Assumptions Housing Economic Analysis Table 23 Development Cost Assumptions and Budgets 63 Table 24 Net Operating Income from Market-Rate Apartments 100% Market Rate Units 64 Table 25 Net Sales Proceeds from Market-Rate Owner Housing 100% Market Rate Units 65 Table 26 Utility Allowances 66 Table 27 Supportable Mortgage Calculations, Affordable Rental High-Rise Prototypes 67 Housing Table 28 Supportable Mortgage Calculations, Affordable Rental Mid-Rise Prototypes Housing 68 Final Report March 27, 2016

14 List of Tables Table Number Table Title Table Subheading Page Number Table 29 Supportable Mortgage Calculations, Affordable Rental Low-Rise Prototypes 69 Housing Table 30 Affordable Home Sales Price Calculations, Owner Prototype 5 70 Housing Table 31 Affordable Home Sales Price Calculations, Owner Prototype 7 71 Housing Table 32 Affordable Units by Prototype and Income Level Rental Housing 72 Inclusionary Scenarios Table 33 Affordable Units by Prototype and Income Level Owner Housing 73 Inclusionary Scenarios Table 34 Market-Rate Units by Prototype Inclusionary Housing 74 Scenarios Table 35 Affordable Unit Square Feet by Prototype and Income Inclusionary Housing 75 Level Scenarios Table 36 Market-Rate Unit Square Feet by Prototype Inclusionary Housing 76 Scenarios Table 37 Rental Housing Income from Inclusionary Units Rental Housing 77 Inclusionary Scenarios Table 38 Total Rental Housing Income by Scenario Rental Housing 78 Inclusionary Scenarios Table 39 Apartment NOI by Scenario Rental Housing 79 Inclusionary Scenarios Table 40 For-Sale Housing Sales Proceeds from Inclusionary Owner Housing 80 Units Inclusionary Scenarios Table 41 Total Net Sales Proceeds for Owner Housing by Owner Housing 81 Scenario Inclusionary Scenarios Table 42 Return on Equity Analysis Residential Development Low Cap Rate 82 Prototypes Assumption Table 43 Return on Equity Analysis Residential Development High Cap Rate 83 Prototypes Assumption Table 44 Land Residual Analysis Residential Development Low Cap Rate 84 Prototypes Assumption Table 45 Land Residual Analysis Residential Development High Cap Rate 85 Prototypes Assumption Table 46 For-Sale Housing Sales Proceeds from Inclusionary Owner Housing Alternative Set 86 Units Inclusionary Scenarios Aside Scenarios Table 47 Total Net Sales Proceeds for Owner Housing by Scenario Owner Housing Inclusionary Scenarios Alternative Set Aside Scenarios 87 Final Report March 27, 2016

15 1 Executive Summary and Introduction The City of Cambridge Community Development Department ( City ) retained DRA to prepare a study to evaluate the impact of new market-rate residential development on housing affordability and socio-economic diversity in Cambridge and to recommend changes, if any, to the inclusionary housing provisions of Article of the Cambridge Zoning Ordinance ( the Zoning Ordinance and the City s Inclusionary Housing Program to better meet the City s policy goals for the program. DRA evaluated the City s current Inclusionary Housing Program (IHP) in three major respects: 1. In terms of housing and demographic changes since the initial inclusionary housing studies were completed for the City in 1997 and 1998; 2. Through an economic analysis that examines the effect of alternative inclusionary housing standards on residential financial feasibility and land values in Cambridge using a series of prototypical housing developments; and 3. In light of best practices in inclusionary housing programs as informed by DRA s extensive nationwide experience as well as a survey of selected inclusionary housing programs across the country. The study includes an overview of demographic and residential market conditions in Cambridge, with a focus on trends in market housing prices and rents, housing affordability, household income distribution, housing cost burden, and the affordable housing stock. Using this data, DRA evaluates changes in socioeconomic diversity in the City since Final Report 1

16 DRA s economic analysis evaluates the effect of the City s current inclusionary housing program, and potential changes to the program, on the financial feasibility of new residential development in Cambridge. DRA worked closely with City staff to develop a series of residential prototypes that represent the types of projects currently being built in Cambridge and reflect current underlying zoning designations in the City. The prototypes include large high- and mid-rise rental housing prototypes as well as smaller low- and mid-rise homeownership and rental developments. These prototypes form the basis of DRA s economic analysis of the current program and alternative set-aside and income targeting policies. The findings of the analysis will assist the City in evaluating policy options for the inclusionary housing program that will generate affordable housing units to meet needs in the community while being sensitive to current and future real estate market conditions. Summary of Key Findings The City of Cambridge has a population of approximately 109,700 (July 1, 2014) with about 46,000 households, of which 65% are renters and 35% are owners. Increases in residential market rents and sales prices have outstripped increases in area median income (AMI) in the City of Cambridge since 1997, resulting in a marked reduction in housing affordability, an increase in costburdened households, and a decline in the proportion of households in the City with incomes under 100% of AMI. The City s Inclusionary Housing Program, along with the Cambridge Affordable Housing Trust and other City programs to develop and preserve affordable housing, have succeeded in maintaining the proportion of the restricted affordable housing stock at approximately 15% of total housing units. As the development of new affordable housing becomes more challenging due to market competition for sites and declining state and federal funding for affordable housing, the Inclusionary Housing Program is contributing an increasing proportion of new affordable units in the City. Thus, the continuation and strengthening of the City s Inclusionary Housing Program is vital to maintaining the affordable housing stock and preserving socioeconomic diversity within the community. As currently configured the Final Report 2

17 Inclusionary Housing Program cannot on its own maintain the existing proportion of affordable housing. Overall, the analyses demonstrate that there is room for Cambridge to increase its inclusionary standard without rendering housing development economically problematic. The higher the inclusionary housing standard and the deeper the affordability, the greater the impact on the feasibility of residential development. As the City assesses options, these analyses are useful to consider so as not to either discourage development or tilt development in favor of commercial projects. Final Report 3

18 4 Contents of Report This report is presented in the following major sections: n Existing Inclusionary Housing Program (IHP) The first section reviews the background and history of the Cambridge inclusionary housing program, along with a summary of key provisions of the ordinance. It also compares several of these key provisions with other selected inclusionary zoning programs nationwide, and presents available data on affordable unit production in Cambridge and the other selected communities. n Socioeconomic Diversity Analysis The second section of the report provides an analysis of trends in socioeconomic diversity in the City of Cambridge, based on a review of demographic and market data. The analysis includes a review of market-rate rents and sales prices, household size, housing unit bedroom count distribution, housing affordability, household income distribution and the affordable housing stock in the City. n Economic Analysis of IHP Standards and Options The third section of the report presents an economic feasibility analysis of the impact of the existing Inclusionary Housing Program and alternative program setaside standards on the economic feasibility of new rental and owner housing development in Cambridge. The analysis uses a series of housing prototypes representing different renter and owner housing products currently being built in Cambridge. n Policy Options and Recommendations The final section of the report reviews policy issues of concern to the City of Cambridge regarding the inclusionary housing provisions of the Zoning Ordinance, as well as several additional issues identified by DRA in its review of the City s program. DRA describes policy options for these issues and makes policy recommendations for the City s consideration. These options and recommendations are based on the socioeconomic diversity analysis, the economic feasibility assessment, the review of selected inclusionary housing programs in Final Report 4

19 other communities, and DRA s experience with inclusionary housing best practices nationwide. Detailed Summary of Findings Findings of Socioeconomic Diversity Analysis DRA analyzed trends in market-rate housing rents and sales prices, the affordability of housing, household income distribution, household size and unit bedroom count and the affordable housing stock, to determine how recent market trends and conditions have affected the socioeconomic diversity of Cambridge. Key Findings: Increases in residential market rents and sales prices have outstripped increases in area median income (AMI) in the City of Cambridge since 1997, resulting in a marked reduction in housing affordability, an increase in cost-burdened households, and a decline in the proportion of households in the City with incomes under 100% of AMI. The City s Inclusionary Housing Program, along with the Cambridge Affordable Housing Trust and other City programs to develop and preserve affordable housing, have succeeded in maintaining the proportion of the restricted affordable housing stock at approximately 15% of total housing units. As the development of new affordable housing becomes more challenging due to market competition for sites and declining state and federal funding for affordable housing, the Inclusionary Housing Program is contributing an increasing proportion of new affordable units in the City. Thus, the continuation and strengthening of the City s Inclusionary Housing Program is vital to maintaining the affordable housing stock and preserving socioeconomic diversity within the community. As currently configured the Inclusionary Housing Program cannot on its own maintain the existing proportion of affordable housing. These findings are based on the following key observations: Final Report 5

20 1. Between 2007 and 2014, area median income increased a total of 14%, while the median two-bedroom asking rent increased 31%, the average condo sales price increased 33%, and the average single-family sales price increased 45%. 2. Housing affordability has declined markedly in Cambridge since the inception of the inclusionary zoning program. In 1997, the affordable rent at 80% of AMI was almost equal to the median asking market rent, with a ratio of affordable to market rent of 99.3%. By 2000, the percentage declined to 72.1%, and by 2014 the percentage was 63.8%. For owner housing, prices fell modestly during the recession but have increased in recent years and are generally out of reach for households earning less than 100% AMI. The percentage of AMI required to afford the average-priced condo in the City increased from 145% in 2007 to 151% in The percentage of AMI required to afford the average-priced single-family home increased from 226% in 2007 to 257% in Households are considered cost-burdened if they pay more than 30% of their gross income on housing, and severely cost-burdened if they pay more than 50% of their income on housing. Housing cost burden has increased in Cambridge over the 2000 to 2011 period. In 2000, approximately 40% of all renter households paid more than 30% of their gross income on housing, and 19% paid more than 50% of their income on housing. The most recently available CHAS data for 2011 show that the percentage of cost-burdened renter households increased to 45%, and 24% were severely cost-burdened. The percentages of cost-burdened households are much higher for households with incomes below 80% AMI. For example, more than 56% of households with incomes between 50% and 80% of AMI paid more than 30% of their income on housing in 2000, rising to 74% in Based on the most recent demographic data available, between 2000 and 2011 the percentage of total households with incomes below 50% remained relatively stable but the percent of households with incomes between 50% and 100% of AMI declined from 27% to 18%. A decline is seen in both renter and owner households in these income groups, and has likely only accelerated since 2012, given the continued rapid increase in housing costs. Over the same 2000 to 2011 period, the percentage of households with incomes above 120% of AMI increased from 35% to 47%. 5. The affordable housing stock in Cambridge has remained fairly constant at 15% of total housing units since the inception of the City s Inclusionary Final Report 6

21 Housing Program in However, the percentage of the City s affordable housing stock comprised of inclusionary units has increased from 6% in 2008 to almost 11% in 2014, demonstrating the increased importance of the City s inclusionary zoning program in creating new affordable housing in Cambridge. The findings of the socioeconomic diversity analysis indicate that households with incomes between 50% and 80% of AMI have been particularly hard hit by changing market conditions. This is demonstrated by the lack of market-rate units affordable to this group, the large increase in cost-burdened households, and the decline in the proportion of households in this income category. These findings support the continued targeting of the inclusionary housing program at the 65% AMI level, particularly for renters. The findings indicate that the 80% to 100% of AMI group is another segment of the population that is being squeezed out by rising housing costs in Cambridge. DRA also reviewed trends in household size and housing unit bedroom count distribution, to determine whether a change to the inclusionary zoning program to incentivize, or require, the creation of larger family-sized units is warranted. In regard to household size, there appears to be a shift toward two-person renter households relative to both smaller and larger households. Based on available data on bedroom count for the entire housing stock, it is difficult to see any trend towards smaller units emerging. But in the larger new developments there is a trend toward smaller units, resulting in smaller inclusionary units. Additionally, as in many communities, low-income large family housing needs continue to be unmet as market prices for family sized units continue to outstrip income. This may contribute to the perception of increased scarcity of family-sized units. Findings of Economic Feasibility Analysis DRA prepared an economic analysis to assist the City in evaluating potential revisions to its Inclusionary Housing Program for residential development. DRA conducted the economic analysis using seven housing prototypes, developed in conjunction with City staff. These prototypes represent the type of housing developments that have been recently developed and are in the development pipeline in the City. Final Report 7

22 Key Findings: Cambridge, within limits, can increase the percentage of units set aside and alter the income targeting of its IHP without rendering market-rate housing economically problematic. The higher the inclusionary housing standard and the deeper the affordability, the greater the impact on the feasibility of residential development. As the City assesses options, these analyses are useful to consider so as not to either discourage development or tilt development in favor of commercial projects. The prototypes analyzed include low-, mid- and high-rise residential developments appropriate to a range of current zoning designations in the City. The tenure, number of units and building heights of the prototypes are summarized in Table 1. Table 1 Summary of Development Prototypes Cambridge Inclusionary Housing Study Prototype Tenure Total Units Appx. Bldg. Stories Density (DUs/Acre) Prototype 1 Rental Prototype 2 Rental Prototype 3 Rental Prototype 4 Rental Prototype 5 Owner Prototype 6 Rental Prototype 7 Owner Source: City of Cambridge; DRA. DRA interviewed a number of residential and mixed-use developers active in Cambridge to review the prototypes, revenue, operating cost and development cost assumptions used in the analysis. DRA incorporated comments received from these developers into the analysis. DRA also consulted with the City s Assessing Department regarding current cap rates, rates of return on equity, and the ratio of debt to equity on recent residential development projects in the City of Cambridge. Final Report 8

23 DRA evaluated the economic feasibility of the prototype developments using both Return on Equity (ROE) and Residual Land Value (RLV) analyses. The return on equity approach calculates the value of a development based on its stabilized income potential and subtracts the costs of development (including land) to determine the net value of the investment, or developer/investor profit. Under the ROE approach, the financial feasibility of the prototypes is measured by the rate of the return on equity that the resulting net investment value (or profit) represents. Land costs are held fixed at an estimated market land price and the economic impact of the program is shown as a change in the dollar amount of the net value of investment in the prototype and as a change in the ROE. Land residual analysis methodology calculates the value of a development based on its income potential and subtracts the costs of development (excluding land but including an assumed return on equity), to yield the underlying value of the land. An alternative that generates a value to the land that is negative, or well below market land sales prices, is financially problematic. Both the ROE and RLV analyses calculate the value of rental prototypes at a point in time based on the estimated stabilized net operating income of the prototype. The economic performance of the prototypes was calculated for the following four set-aside scenarios, which are the same for both the renter and owner prototypes. The prototypes assume the density bonus of market rate units is already incorporated. Scenario 1: 11.5% of total units at 65% of AMI (typical project under current IHP standards) Scenario 2: 15% of total units at 65% of AMI plus 5% of units at 100% of AMI. Scenario 3: 20% of total units at 65% of AMI. Scenario 4: 20% of total units at 65% of AMI plus 5% of units at 100% of AMI. DRA also examined 20% set-asides for owners at average AMIs of 75% and 90%. The analysis was conducted under two alternative capitalization ( cap ) rate assumptions. Based on consultation with the City Assessing Department, current cap rates in Cambridge are approximately 4.00% for projects with 50 units or more and 4.25% for projects with 50 units or less. These current cap rates are at historically low levels. We compare the results of the economic analysis using Final Report 9

24 current historically low cap rates (estimated at 4.0%) to higher rates (assumed at 4.75%). The higher cap rates reflect a prospective economic view. An acceptable ROE is estimated at 8.0% or more. The results of the analysis are designed to inform the city as it evaluates potential changes to the program and are one among other factors to consider in the public policy decision. The results of the ROE and RLV analyses are summarized in Table 2 under the low cap rate scenario and Table 3 under the higher cap rate scenario. The findings are summarized as follows. Overall, the analyses demonstrate that there is room for Cambridge to expand its inclusionary housing standards without rendering housing development economically problematic. The higher the standards, the greater the impact on the feasibility of residential development. As the City reviews options, these analyses are useful to consider so as not to either discourage development or tilt development in favor of commercial projects. 1. For the rental prototypes under the low cap rate assumption, the ROE for all of the prototypes substantially exceeds the acceptable rate of return threshold of 8% for Scenarios 1, 2 and 3. The ROE also equals or exceeds the threshold under Scenario For the rental prototypes under the higher cap rate assumption, the ROE exceeds the acceptable rate of return for all rental prototypes for 100% marketrate development. Under Scenario 1 (the existing program), the ROE exceeds the acceptable rate of return for Prototypes 2, 3 and 6, and falls slightly below the acceptable rate of return for Prototypes 1 and 4. It also exceeds the acceptable rate of return under Scenarios 2 and 3 for Prototypes 3 and 6. The ROE falls below the acceptable rate of return for all prototypes except Prototype 6 under Scenario For owner Prototype 5, the ROE exceeds the acceptable rate of return for the market-rate development and Scenario 1 (the existing program). It drops slightly below the acceptable rate of return to 7% for Scenarios 2 and 3, and further below the acceptable rate of return under Scenario 4. DRA also examined an owner housing set-aside of 20% at average income levels of 75% and 90% AMI. The ROE meets or exceeds the acceptable rate of return under both of these alternatives. Final Report 10

25 4. For the small six-unit owner Prototype 7, the ROE exceeds the acceptable rate of return for the 100% market-rate development and all Scenarios examined (assuming one affordable unit, or 17% of total units, in each case). 5. The RLV analysis shows similar results. Those scenarios with ROEs in excess of the acceptable rate of return generate land values in excess of the assumed land sales prices. Those that fall below the acceptable rate of return also generate land values less than the assumed land sales prices, in some cases substantially less. Regulation and development impact fees on residential development that increase the costs of development, including inclusionary housing standards, will ultimately be passed through to the land owner in the form of reduced land prices. In order for developers to profitably develop new housing, they will bid down land prices to the level that makes development feasible, given market economics and zoning regulations affecting the amount of development that can be built on a particular site. This also depends if the site can be developed with commercial as well as residential uses. Increasing the cost of residential development relative to the cost of commercial development may change the highest and best use in favor of commercial development and prevent the land value from being bid down. In addition, land prices react more quickly to factors that increase land prices, such as increases in rents. Land prices tend to be slower to respond to factors that decrease land prices, including changing market conditions and increased regulation or fees, as owners who purchased recently may be reluctant to take a loss and others may be hesitant to adjust their expectations downward. Land prices are also volatile in response to economic cycles and factors beyond the control of local government. Land will lose value in higher cap rate environments. If the residual land value is negative, that indicates that capitalized values are not sufficient to cover the other development costs besides land, and new development will be slowed or halted until market conditions change. Therefore, very low or negative RLVs such as shown under Scenario 4 for rental Prototypes 1 and 4 with the higher cap rate assumption suggest that development of these project types would not occur until market conditions change. Final Report 11

26 Table 2 Resid. Cap Rate Summary of Return on Equity and Land Residual Analysis Results Under 50 Units 4.25% Residential Development Prototypes 50 Units or More 4.00% Low Cap Rate Assumption Prototype 1 Prototype 2 Prototype 3 Prototype 4 Prototype 5 Prototype 6 Prototype 7 Tenure Rental Rental Rental Rental Homeownership Rental Homeownership Residential Units Site Area (SF) 53, ,974 53,033 10,026 23,791 6,800 6,800 Residential Net SF 207, ,250 79,550 19,550 43,300 6,800 6,800 Total Net SF 207, ,250 79,550 19,550 43,300 6,800 6,800 Residential Units Parking Spaces Approximate Building Stories Assumed Land Price Per Unit $50,000 $85,000 $85,000 $85,000 $85,000 $170,000 $170,000 Per SF $282 $224 $160 $212 $143 $150 $150 Number of Inclusionary Units 100% Market Rate Scenario 1: Existing IHO (1) Scenario 2 (2) Scenario 3 (3) Scenario 4 (4) Return on Equity (ROE) (5) 100% Market Rate 36% 30% 47% 38% 29% 54% 50% Scenario 1: Existing IHO (1) 25% 21% 36% 25% 16% 40% 34% Scenario 2 (2) 18% 17% 29% 20% 10% 40% 34% Scenario 3 (3) 17% 14% 27% 16% 9% 39% 34% Scenario 4 (4) 11% 11% 23% 13% 5% 40% 34% Residual Land Value (RLV) (6) 100% Market Rate Per Unit $206,180 $193,471 $315,214 $220,743 $202,273 $395,526 $386,064 Per SF $1,161 $509 $594 $550 $340 $349 $341 Scenario 1: Existing IHO (1) Per Unit $145,372 $155,115 $254,952 $162,916 $134,913 $329,344 $306,122 Per SF $819 $408 $481 $406 $227 $291 $270 Scenario 2 (2) Per Unit $110,966 $136,606 $216,326 $142,719 $101,860 $329,344 $306,122 Per SF $625 $360 $408 $356 $171 $291 $270 Scenario 3 (3) Per Unit $101,797 $122,694 $202,553 $126,314 $96,661 $326,125 $306,122 Per SF $573 $323 $382 $315 $163 $288 $270 Scenario 4 (4) Per Unit $69,631 $107,770 $178,004 $110,836 $77,310 $328,057 $306,122 Per SF $392 $284 $336 $276 $130 $289 $270 Assumed Return on Equity (7) 8% 8% 8% 8% 8% 8% 8% Assumed Investment Period (Years) (1) 11.5% of total units at 65% of AMI for renters and owners. (2) 15.0% of total units at 65% of AMI plus 5% of units at 100% AMI for renters and owners. (3) 20.0% of total units at 65% of AMI for renters and owners. (4) 20% of units at 65% of AMI plus 5% of units at 100% of AMI for renters and owners. (5) Return on equity measured as net project value divided by the number of years equity investment divided by total equity investment. (6) Land residual value per housing unit and per square foot site area. (7) Used in land residual analysis. Source: DRA.

27 Table 3 Resid. Cap Rate Summary of Return on Equity and Land Residual Analysis Results Under 50 Units 5.00% Residential Development Prototypes 50 Units or More 4.75% High Cap Rate Assumption Prototype 1 Prototype 2 Prototype 3 Prototype 4 Prototype 5 Prototype 6 Prototype 7 Tenure Rental Rental Rental Rental Homeownership Rental Homeownership Residential Units Site Area (SF) 53, ,974 53,033 10,026 23,791 6,800 6,800 Residential Net SF 207, ,250 79,550 19,550 43,300 6,800 6,800 Total Net SF 207, ,250 79,550 19,550 43,300 6,800 6,800 Residential Units Parking Spaces Approximate Building Stories Assumed Land Price Per Unit $50,000 $85,000 $85,000 $85,000 $85,000 $170,000 $170,000 Per SF $282 $224 $160 $212 $143 $150 $150 Number of Inclusionary Units 100% Market Rate Scenario 1: Existing IHO (1) Scenario 2 (2) Scenario 3 (3) Scenario 4 (4) Return on Equity (ROE) (5) 100% Market Rate 17% 12% 26% 16% 29% 29% 50% Scenario 1: Existing IHO (1) 8% 5% 17% 4% 16% 17% 34% Scenario 2 (2) 2% 1% 12% 0% 10% 17% 34% Scenario 3 (3) 1% -1% 10% -3% 9% 17% 34% Scenario 4 (4) -4% -4% 6% -6% 5% 17% 34% Owner: 75% AMI 10% 35% Owner: 90% AMI 12% 37% Residual Land Value (RLV) (6) 100% Market Rate Per Unit $103,567 $111,095 $197,284 $125,338 $202,273 $281,244 $386,064 Per SF $1,161 $509 $594 $550 $340 $349 $341 Scenario 1: Existing IHO (1) Per Unit $52,360 $78,796 $146,536 $76,185 $134,913 $224,989 $306,122 Per SF $295 $207 $276 $190 $227 $199 $270 Scenario 2 (2) Per Unit $23,387 $63,210 $114,009 $59,017 $101,860 $224,989 $306,122 Per SF $132 $166 $215 $147 $171 $199 $270 Scenario 3 (3) Per Unit $15,666 $51,494 $102,410 $45,073 $96,661 $222,253 $306,122 Per SF $88 $136 $193 $112 $163 $196 $270 Scenario 4 (4) Per Unit ($11,422) $38,926 $81,738 $31,917 $77,310 $223,895 $306,122 Per SF ($64) $102 $154 $80 $130 $198 $270 Owner: 75% AMI Per Unit $102,789 $311,378 Per SF $173 $275 Owner: 90% AMI $111,954 $319,247 Per Unit $188 $282 Per SF Assumed Yield on Equity (7) 8% 8% 8% 8% 8% 8% 8% Assumed Investment Period (Years) (1) 11.5% of total units at 65% of AMI for renters and owners. (2) 15.0% of total units at 65% of AMI plus 5% of units at 100% AMI for renters and owners. (3) 20.0% of total units at 65% of AMI for renters and owners. (4) 20% of units at 65% of AMI plus 5% of units at 100% of AMI for renters and owners. (5) Return on equity measured as net project value divided by the number of years equity investment divided by total equity investment. (6) Land residual value per housing unit and per square foot site area. (7) Used in land residual analysis. Source: DRA.

28 Existing Inclusionary Housing Program Background and History The City of Cambridge adopted the Inclusionary Zoning Ordinance in 1998, in large part as a response to the impact the end of rent control in 1995 was having on the City s housing market. 1 In 1997 and 1998, the City of Cambridge contracted with consultant teams to conduct studies on rationales and approaches for adoption of inclusionary housing provisions in the Zoning Ordinance. The 1997 study by Peter Werwath and Associates (Recommendations Concerning a New Inclusionary Zoning Ordinance, A Report to the City of Cambridge) recommended a framework and provisions for inclusionary housing standards to be included in the Zoning Ordinance. The 1998 study by Stockard & Engler & Brigham (Cambridge Inclusionary Housing Study) explored affordability standards for the inclusionary provisions in the Zoning Ordinance. The 1998 study concluded that for every 10 new market rate housing units, 1.5 affordable units were needed to maintain the proportion of affordable housing units in the City and assist in meeting the City s socio-economic diversity goals. Based on these studies, the City adopted the inclusionary housing provisions in the Zoning Ordinance in The Zoning Ordinance asks developers of residential projects of 10 or more units or 10,000 or more square feet to mitigate the impact of their development on the socio-economic diversity of Cambridge by providing 15% of the units as affordable, and provides such developers with a development bonus increasing the FAR normally permitted in the applicable zoning district by 30%, half of which must be allocated to the affordable units. When aggregating all projects, the result is an overall ratio of approximately 12.5% affordable units to total units built. Since its inception through the end of 2015, the Inclusionary Housing Program in Cambridge has been responsible for the creation of nearly 850 affordable housing units, built or underway, including more than 650 rental units and over Rent control was ended in the City due to a statewide referendum in 1995 that eliminated rent control across Massachusetts.) Final Report 14

29 homeownership units located in privately owned buildings throughout the City 1. Approximately 50% of inclusionary rental units are filled with tenants with mobile housing vouchers. The City has been able to maintain the ratio of 1.5 affordable units for every 10 market rate units, as set forth in the original inclusionary zoning studies, through employment of a number of strategies, including inclusionary zoning. In addition to inclusionary zoning, affordable housing is created in Cambridge through assisted housing development, homeownership programs run by the City, the preservation of expiring use projects, and public housing and mobile vouchers from the Cambridge Housing Authority. The Affordable Housing Trust funds non-profit and other affordable housing developers to produce 100% affordable projects, which are largely targeted to lower income households. Trust funded projects leverage additional state and federal resources, including the federal 9% Low Income Housing Tax Credit program (9% tax credits); the most valuable source of leveraged financing available today. Inclusionary Housing Provisions of the Cambridge Zoning Ordinance The following sections describe the key provisions of the existing inclusionary housing provisions in the Zoning Ordinance and the Inclusionary Housing Program. 1. Income Targeting, and Housing Payment Standards Eligible households must earn between 50% and 80% of the area median income (AMI), with a target income of 65% of the AMI. Households who have access to a mobile housing voucher may have a lower income. Rent for inclusionary units are set for each tenant at 30% of the actual gross income of the tenant household. A utility allowance, based on the Cambridge Housing Authority schedule, is subtracted for any utilities that the tenant will pay directly. For homeownership units, sales prices are set so the affordable housing expense (including mortgage principal and interest, property insurance, real estate taxes, 1 These figures include several developments that are subject to other similar inclusionary zoning like requirements, such as the recent Alexandria development in Kendall Square. Final Report 15

30 condo/hoa fees and parking fees, where applicable) does not exceed 30% of 65% of the AMI. 2. Unit Location Cambridge requires all inclusionary units to be provided on site, unless the project can demonstrate to the Planning Board that providing the units on site would create a significant hardship because the property cannot physically accommodate the required affordable units. All units have been provided on site. No developers have sought an exemption. 3. Set Aside Percentages and Density Bonus Provisions Inclusionary zoning applies to new or converted residential projects with 10 or more units or with over 10,000 square feet of residential space. Fifteen percent of the dwelling units, before the bonus is applied, are to be affordable. The FAR normally permitted in the applicable zoning district for residential uses is increased by 30% for inclusionary projects, of which at least 50% of the additional FAR must be allocated to the affordable units. When the bonus is applied, the 15% requirement results in approximately 11.5% of the total units as affordable. 4. Unit Comparability Affordable units in an inclusionary project must be generally comparable in size and materials to the other units in the overall project. The City also requires the affordable units to be distributed proportionately throughout the building in terms of floors, the direction they face, number of bedrooms, and other factors. Specific units are designated as affordable and remain so for the life of the project. 5. Parking One parking space must be provided for each affordable unit unless there is an average of less than one space per unit for the entire development, in which case the parking spaces provided for the affordable units shall be in the same proportion as for the market-rate units. If there is a parking fee for the market rate units, there is a reduced parking fee for the inclusionary units. 6. Administration Prior to the issuance of a Building Permit for a property, a permanent deed restriction (the Inclusionary Housing Covenant) is required to be recorded. The Final Report 16

31 Covenant identifies the inclusionary units and the other elements of the program, and is the senior interest on the property. When construction is complete, the City manages tenant and homeowner selection. The City Community Development Department (CDD) maintains a single point of entry for applicants to apply for affordable inclusionary rental and homeownership units. CDD conducts the marketing and maintains applicant lists. Applicants are screened by the City for eligibility and forwarded for final approvals to property managers for rental units and for mortgage approvals for homeownership units. Inclusionary Housing Programs in Selected Communities DRA reviewed the provisions of inclusionary housing programs in selected communities nationwide, with a focus on policy issues of interest to Cambridge, as well as available data on the number of affordable units produced under these programs. DRA consulted with City staff to select the communities below for analysis which represent municipalities similar to Cambridge, and/or which have inclusionary programs that include features of relevance to Cambridge housing policy goals. Appendix B includes inclusionary housing program profiles for the following communities: Boston, MA Boulder, CO Burlington, VT Montgomery County, MD San Francisco, CA Santa Monica, CA Somerville, MA Washington, DC Table 4 compares the percentage of affordable units, density bonus provisions, threshold project sizes and affordable housing unit production in Cambridge with the inclusionary housing programs surveyed. Somerville s existing program is included, but the city has proposed changes to its program that are currently being considered. The current population and recent population growth rate for each community is shown for context. Final Report 17

32 City Year Adopted Table 4 Inclusionary Housing Set-Aside, Density Bonus, and Threshold Size Provisions and Affordable Units Created Selected Nationwide Inclusionary Housing Programs 2015 % of Units or FAR Rental % of AMI Cambridge, % 50%-80% MA 2 AMI Boston, MA On-Site % of Units or FAR % 70% AMI 6.5% plus 6.5% 4 Owner % of AMI Density Bonus Project Size Threshold 15% Up to 80% 30% 10 units or 80% AMI 100% AMI Off-Site 15-18% 3 70% AMI 15-18%4 80%-100% AMI 4 10,000 SF Afford. Units Created Population ( % Pop. Change) ,355 (4.9%) Negotiated 10 units 5 1, ,966 (9.4%) Boulder, CO % 60% AMI 20% 76% AMI 8 None All projects ,166 (9.1%) Burlington, VT % to 25% 65% 15% to 25% 75% 15% to 25% 5 units ,284 (6.1%) Montgomery Co., MD % to 15% 65% AMI (70% AMI hi-rise) 12.5% to 15% None 70% Up to 20 to 22% 20 units 370 Ave. Annual 1,016,677 (15.9%) 1 The time period over which the number of affordable units created varies from community to community based on available data; therefore these figures are not directly comparable between programs but are intended to give a sense of whether the program has been effective in creating affordable units. 2 Program requires 15% of pre-bonus units to be affordable at actual household income for renters (with a target average of 65% AMI) and at 65% of AMI for owners. With the 30% FAR bonus, the set-aside is equivalent to 11.5% of total post-bonus units. The number of units created is since inception in 1998 and includes units created through inclusionary zoning and additional zoning provisions similar to inclusionary zoning. 3 For renters, 13% on-site and 15%-18% offsite requirement at 70% AMI (under certain circumstances a portion of units may be designated up to 100% AMI). 4 For owners, 13% on-site and 15%-18% offsite requirement including at least 50% of units at 80% AMI and the balance of units at 100% AMI. 5 Applies to residential developments of 10 or more that seek zoning relief, or receive city financing, or land owned by the city or BRA. 6 Approximate units completed or in construction Most affordable rental units have been developed off-site. For-sale developments should provide at least one-half of required units on-site; others may be provided on-site or through cash-in-lieu contribution 8 HUD low-income limit plus 10%, currently about 76% AMI The affordable housing set-aside percentage depends on the affordability of average price/rent of the market-rate units. The set-aside is 15% if it is less than 13% of median income; 20% if it is 1450% to 179% of AMI, and 25% is if it is 180% of AMI or above. Developers may provide affordable units off-site at 125% of the on-site obligation March 2014.

33 Table 4 (Continued) Inclusionary Housing Set-Aside, Density Bonus, and Threshold Size Provisions and Affordable Units Created Selected Nationwide Inclusionary Housing Programs 2015 City Year Adopted % of Units or FAR Rental % of AMI % of Units or FAR Owner % of AMI Density Bonus Project Size Threshold Afford. Units Created Population ( % Pop. Change) San Francisco, CA Citywide Onsite % 55% AMI 12% 90% AMI None 10 units 1, ,442 (7.7%) Citywide Offsite 20% 17% hi-rise 55% AMI 20% 70% AMI None Plan Areas Onsite 14.4% to 17.6% 55% AMI 14.4% to 17.6% 90% AMI Plan Areas Offsite Santa Monica, CA % or 10% or 20% 23% to 27% 55% AMI 23% to 27% 70% AMI None 30% AMI 50% AMI 80% AMI Somerville, MA % At least ½ at 50% AMI; balance at 80% Washington, % of FAR DC. 18 plus 5% of FAR 50% AMI 80% AMI 20% (4-15 units) 25% (16+ units) 12.5% At least ½ at 80% AMI; balance at 110% 5% of FAR plus 5% of FAR 100% AMI 5% to 35% 2 units 1, ,472 (3.2%) 50% AMI 80% AMI Source: DRA survey of selected inclusionary housing programs. Note: % of AMI is for eligibility. Two additional market-rate units for each additional affordable unit 16 8 units ,804 (1.7%) Up to 20% 10 units ,449 (13%) 12 The time period over which the number of affordable units created varies from community to community based on available data; therefore these figures are not directly comparable between programs but are intended to give a sense of whether the program has been effective in creating affordable units March Rental project set-asides are lower if the developer chooses lower income levels, as shown, and are the same for on- and off-site units. For owner hosing, set-aside requirements vary by project size and are 25% higher for off-site compliance. The amount of the density bonus depends on the percentage of units that are affordable and the income level, with higher density bonuses given for lower AMIs and higher set-aside percentages. 15 Approximate production 1998-March If more than 12.5% affordable units are provided, may apply for a bonus of 2 market rate units for each affordable unit, subject to conditions and restrictions present. 18 Set-aside varies by construction type and zone district. Generally in low density zones the set-aside is the greater of 10% of residential FAR or 75% of bonus density. In higher density zones it is the greater of 8% of FAR or 50% of the bonus density split evenly between 50% AMI and 80% AMI March 2014.

34 1. Comparison of Inclusionary Housing Program Provisions In the communities surveyed, income targeting for rental and ownership housing typically differ, with higher income targeting for owner than for renter housing. This policy is often dictated by market prices. In most cities, Cambridge included, rental housing is more affordable than ownership housing. For rental housing, the programs in the other communities surveyed target households earning below 80% of AMI, with income targets from 30% of AMI (as an option in Santa Monica, at a lower percentage of units), to 80% of AMI. Except for Washington D.C., which has identical income targets for owners and renters like Cambridge, all of the surveyed programs target higher income levels for owners than renters. The income targets for owners range from 70% to 100% of AMI. Among the communities surveyed, lower income targeting is associated with a lower set-aside percentage, while higher income targeting is associated with a higher percentage of affordable units. In terms of affordable housing set-aside percentages, Cambridge s current effective requirement of 11.5% of total units at an average of 65% of AMI is in the middle of the pack compared with the communities surveyed. Its percentage of units for owners is also towards the middle of the range of communities surveyed (though at lower income targeting). Of the seven communities surveyed, four offer fixed density bonuses, two offer none, and the seventh (Boston) offers them on a negotiated basis. The density bonus of 30% offered by Cambridge is higher than all but Santa Monica s. Santa Monica requires 11% of units at 50% of AMI or 20% of units at 80% of AMI to qualify for the maximum density bonus of 35%. Unit size thresholds for the inclusionary programs range from 2 units in Santa Monica to 20 units in Montgomery County, Maryland. Boulder has no size threshold. Boston, San Francisco and Washington DC have a 10-unit threshold and Burlington, Vermont has a 5-unit threshold. A comparison of Cambridge s ordinance to the other programs evaluated reveals a number of key differences: Cambridge s ordinance is unique in terms of providing virtually no option for alternative compliance methods such as in lieu fees or offsite compliance. In addition, Cambridge does not allow the affordable units to be smaller units or to have lower quality finishes, and Cambridge specifies that units must be distributed proportionately throughout the building. Cambridge s standard for identical affordable units have advantages and disadvantages. The obvious benefit is that high quality affordable housing is created Final Report 20

35 in mixed-income locations. By not accepting off site units or fees in lieu of units, over 850 affordable units have been created without any public funds and units are completed and occupied at the same time as the market rate units. An additional advantage is that the program is very predictable, and developers are able to determine in advance precisely what will be expected. However, allowing alternative compliance options could result in a lower cost of compliance to developers, which in turn could allow for increased percentages of affordable units that would result in the production of additional units and/or family sized units. Cambridge s program is also unique with respect to its treatment of homeowner association (HOA) dues. Even with a subsidized purchase price, homeownership can be a stretch financially for lower income households with incomes at or below 65% of AMI, especially with respect to maintenance costs and homeowner association (HOA) dues. HOA dues often present a particularly difficult issue for inclusionary housing provisions for owners. Cambridge is unique in that in setting HOA fees, a percentage of interest is determined for each unit and it is applied to HOA fees. The percentage of interest for units is based on value, so the inclusionary units have a lower percentage of interest than market-rate units. As a result, the HOA fees, increases in HOA fees, and assessments are kept lower for affordable units in Cambridge. The challenge presented by HOA dues is one reason many cities allow ownership units to serve a higher income group compared to rental units. 2. Best Practices in Inclusionary Housing Best practices in inclusionary housing relating to policies such as income targeting, density bonus, bedroom size and alternative compliance are contextual to the demographic, market, and land use characteristics of the community as well as to local politics and policy priorities. There is no single best practice in how communities handle these issues. Cambridge s policies, i.e. its practices, will be shaped by policy makers in response to Cambridge s market and demographic characteristics, and by political choices. Key best practices in inclusionary housing include program consistency, clearly documented rules and predictability. Cambridge s program scores very well in this regard. The straightforward and predictable nature of Cambridge s program was repeatedly cited as a positive factor in DRA s interviews with developers involved with the program. Final Report 21

36 22 Socioeconomic Diversity Analysis The City of Cambridge s Inclusionary Housing Program represents an important element in the City s tool kit of resources to assist in the creation and preservation of affordable housing. This section analyzes trends in market-rate housing rents and sales prices, the affordability of housing, household income distribution, and the affordable housing stock. The analysis of recent market trends and conditions demonstrate the ongoing and pressing need for the City s Inclusionary Housing Program. Market conditions in recent years have significantly affected the affordability of housing and the socioeconomic diversity of the community. Key Findings: Increases in residential market rents and sales prices have outstripped increases in area median income (AMI) in the City of Cambridge since 1997, resulting in a marked reduction in housing affordability, an increase in cost-burdened households, and a decline in the proportion of households in the City with incomes under 100% of AMI. The City s Inclusionary Housing Program, along with the Cambridge Affordable Housing Trust and other City programs to develop and preserve affordable housing, have succeeded in maintaining the proportion of the restricted affordable housing stock at approximately 15% of total housing units. As the development of new affordable housing becomes more challenging due to market competition for sites and declining state and federal funding for affordable housing, the inclusionary program is contributing an increasing proportion of new affordable units in the City. Thus, the continuation and strengthening of the City s Inclusionary Housing Program is vital to maintaining the affordable housing stock and preserving socioeconomic diversity within the community. As currently configured the Inclusionary Housing Program cannot on its own maintain the existing proportion of affordable housing. Final Report 22

37 Affordable Housing Income Limits The U.S. Department of Housing and Urban Development (HUD) and most affordable housing programs target specific income categories, defined as a percentage of area median income (AMI) for a given geography. The City of Cambridge is within the Boston-Cambridge-Quincy HUD Metro FMR Area (HMFA) 1, as defined by HUD. HUD publishes the income limits annually adjusted for household size. However, the HUD published 80% AMI for the Cambridge area is artificially capped by a national standard, and is not actually 80% of the 100% of median income. For the City of Cambridge s Inclusionary Housing Program, the City uses the actual calculated 80% of 100% of AMI. 2 The City of Cambridge uses the following terminology in describing these income categories: Low income: households with incomes up to 50% of AMI; Moderate income: households with incomes between 50% and 80% of AMI; Middle income: households with incomes between 80% and 120% of AMI. All of these income limits are based on the 2015 area median income (AMI) of $98,500 for a household of four. Income limits are adjusted by household size using HUD household size adjustment factors. Table 5 shows 2015 income limits for the City of Cambridge Inclusionary Housing Program for household sizes of one to six persons. 1 FMR stands for Fair Market Rent. The Boston-Cambridge-Quincy HMFA is a HUD-defined metropolitan area. 2 HUD s 80% AMI is $69,700 for a four-person household, which is actually 71% of 100% AMI. The calculated 80% AMI used by the Inclusionary Housing Program is $78,800 for a four person household. Final Report 23

38 Table 5 Affordable Housing Income Limits by Household Size City of Cambridge Inclusionary Housing Program 2015 Household Size 50% AMI 80% AMI Actual Calculated 100% AMI 120% AMI One Person $34,500 $55,200 $69,000 $82,800 Two Persons $39,400 $63,040 $78,800 $94,560 Three Persons $44,350 $70,960 $88,700 $106,440 Four Persons $49,250 $78,800 $98,500 $118,200 Five Persons $53,200 $85,120 $106,400 $127,680 Six Persons $57,150 $91,440 $114,300 $137,160 Source: 2015 median household income for the Boston-Cambridge-Quincy HMFA of $98,500, adjusted by household size and income level using HUD household size adjustment factors (actual calculated limits for all categories); DRA. Market-Rate Housing Rents and Sales Prices Cambridge is a high-cost city and the Great Recession had only moderate and temporary effect on rents and sales prices in the City. Twice each year, the City s Community Development Department (CDD) conducts a survey of asking rents to help understand the current rental market from the perspective of a household looking for a rental unit in Cambridge. In recent years, the survey has been conducted using listings from boston.com and craigslist.org. Data sources for prior years depend upon available marketing sources at the time of the survey. Asking market rents best represent the prices facing new households moving to Cambridge or existing residents seeking alternative housing in Cambridge. They are typically higher than average rents paid by existing renter households in the City, as the units on the market may include a disproportionate percentage of newly built units, which typically command higher rents. In addition, in a rising market, rents for existing units are often raised upon turnover. Final Report 24

39 Table 6 summarizes trends in median market-rate asking rents for one-, two- and three-bedroom units in Cambridge since 1997, based on data from the City s annual rent surveys. Table 7 shows the annual compound growth rates in rents for each year, measuring from the prior time period. After increases earlier in the decade, rents declined slightly in and then increased markedly after Overall, from 2010 to 2014 rents increased at annual compound growth rates of 9.5%, 6.4% and 6.4%, for one-, two-, and three-bedroom units, respectively. Table 6 Median Market-Rate Asking Rents City of Cambridge Rent Survey 1997 to BR $1,000 $1,383 $1,800 $1,850 $1,795 $2,300 $2,300 $2,385 $2,583 2 BR $1,200 $1,817 $2,250 $2,470 $2,300 $2,627 $2,800 $3,000 $2,950 3 BR $1,500 $2,000 $2,300 $2,725 $2,650 $2,950 $3,175 $3,200 $3,400 Source: City of Cambridge Community Development Department rent survey. Data is based on the second survey of the year, generally during the months of August or September. The figures for 2008 are from the November, 2007 survey. Table 7 Annual Growth Rate in Median Market-Rate Asking Rents 1 City of Cambridge Rent Survey 2000 to BR 11.4% 3.8% 1.4% -3.0% 28.1% 0.0% 3.7% 8.3% 2 BR 14.8% 3.1% 4.8% -6.9% 14.2% 6.6% 7.1% -1.7% 3 BR 10.1% 2.0% 8.8% -2.8% 11.3% 7.6% 0.8% 6.3% 1 Annual growth rate from prior period (compound rate for multi-year periods). Increases for 2000 are from Source: City of Cambridge Community Development Department rent survey; DRA. Final Report 25

40 Market sales prices for owner-occupied homes and condos have also increased substantially over the past several years. Table 8 shows trends in average sales prices for condominiums and single-family homes from 2007 to 2014, based on data provided by the CDD. Table 9 shows the annual compound growth rates in average prices, measuring from the previous date shown. Overall, between 2007 and 2014, the average condo sales price increased 33% and the average singlefamily sales price increased 45%. Table 8 Average Sales Prices for Single-Family Homes and Condos City of Cambridge 2007 to Condos $416,000 $419,000 $415,000 $424,000 $422,250 $445,500 $510,000 $575,000 Single- Family $648,450 $790,000 $682,500 $760,000 $740,000 $810,750 $858,300 $1,200,000 Source: City of Cambridge Community Development Department; DRA Table 9 Annual Growth Rate in Average Sales Prices for Single-Family Homes and Condos 1 City of Cambridge 2007 to Condos 0.7% -1.0% 2.2% -0.4% 5.5% 14.5% 8.3% Single- Family 21.8% -13.6% 11.4% -2.6% 9.6% 5.9% 9.8% 1 Growth rate from prior year. Source: City of Cambridge Community Development Department; DRA. Housing Affordability Ongoing increases in rents and sales prices have outpaced increases in household income, resulting in a reduction in housing affordability in the City. Table 10 summarizes the area median income for the Boston-Cambridge-Quincy HMFA from 1997 to 2014, along with the growth rate in the AMI from the prior period. Since the affordable rent is based on 30% of AMI, adjusted by household size, affordable rents increase at the same rate as AMI. The table also calculates the Final Report 26

41 affordable rent for a four-person family, and the affordable rent as a percentage of the median market-rate asking rent for a two-bedroom unit in Cambridge. Household income declined slightly between 2011 and 2014, while the median asking rent for a two-bedroom unit increased at a compound annual rate of 3.9% over the same period. With increases in rents and sales prices outstripping increases in income, the affordable rent for a four-person family at 80% of AMI has decreased as a percentage of the median asking rent for a two-bedroom unit. In 1997, the affordable rent at 80% AMI was almost equal to the asking market rent, with a ratio of 99.3%. By 2000, the percentage declined to 72.1%. As a result of the effects of the 2007 to 2009 Great Recession on the housing market, the ratio increased again to 79.8% in 2010, before declining to a low of 62.9% in The ratio increased slightly to 63.8% in Table 10 Trends in Area Median Income and Affordable Rents City of Cambridge 1997 to AMI $59,600 $65,500 $82,400 $91,800 $96,500 $97,800 $94,400 $94,100 % Increase in AMI % 3.3% 3.4% 5.1% 1.3% -3.5% -0.3% 80% AMI Affordable $1,192 $1,310 $1,648 $1,836 $1,930 $1,956 $1,888 $1,882 Rent 2 Affordable Rent as a % of Market 99.3% 72.1% 73.2% 79.8% 73.5% 69.9% 62.9% 63.8% Rent 3 1 Annual growth rate from the prior period (compound annual growth rates for multi-year periods). 2 Affordable rent for a 4-person family earning actual calculated 80% of AMI assuming 30% of annual household income for gross rent (including utilities). 3 Affordable rent for a four-person household earning 80% of AMI as a percentage of the median asking rent for two-bedroom apartments in the City from the City s rent survey. The City s 2014 rent data also included an analysis of the affordability of asking rents to households at various income levels by unit bedroom count. Table 11 shows the percentage of units affordable to a four-person household based on the Final Report 27

42 asking rent survey. The analysis indicates that there are almost no market listings affordable to households at or below 80% of AMI. There continue to be some rental options for households with incomes at 100% of AMI, although the percentage of affordable three-bedroom units is very small. Table 11 Affordability of Median Asking Rents for a Four-Person Household City of Cambridge Rent Survey % of AMI 80% of AMI 100% of AMI 120% of AMI Annual Income $47,050 $67,750 $94,100 $112,920 Affordable Rent 1 $1,176 $1,694 $2,353 $2,823 % Affordable 2-BR Listings 0.0% 0.8% 16.7% 49.6% % Affordable 3-BR Listings 0.0% 0.0% 1.7% 10.2% 1 Based on 30% of gross household income. Source: City of Cambridge Community Development Department; DRA. DRA calculated the income level required to afford new market-rate housing, rental and ownership, in Cambridge. Table 12 shows the annual gross income required to afford the median asking rent for one-, two- and three-bedroom units based on the City s rental survey, assuming 30% of income is spent on rent. It shows that the income required to afford a two-bedroom unit has increased from 81% of AMI in 1997 to 125% of AMI in The required income is even higher when utility allowances are included, as required by the Cambridge IHP and virtually all affordable housing programs. One bedroom units require a lower percent of AMI, while three-bedroom units require a higher percent of AMI (before adjustments for household size). Final Report 28

43 Table 12 Annual Gross Income Required to Afford Median Asking Rents for One-, Two- and Three- Bedroom Units City of Cambridge 1997 to Median Rent, 1 BR Unit 1 $1,000 $1,383 $1,795 $2,300 $2,300 $2,385 $2,583 Gross HH Income Required 2 $40,000 $55,320 $71,800 $92,000 $92,000 $95,400 $103,320 Required Income as a % of AMI 3 67% 84% 78% 95% 94% 101% 110% Median Rent, 2 BR Unit 1 $1,200 $1,817 $2,300 $2,627 $2,800 $3,000 $2,950 Gross HH Income Required 2 $48,000 $72,680 $92,000 $105,080 $112,000 $120,000 $118,000 Required Income as a % of AMI 3 81% 111% 100% 109% 115% 127% 125% Median Rent, 3 BR Unit 1 $1,500 $2,000 $2,650 $2,950 $3,175 $3,200 $3,400 Gross HH Income Required 2 $60,000 $80,000 $106,000 $118,000 $127,000 $128,000 $136,000 Required Income as a % of AMI 3 101% 122% 115% 122% 130% 136% 145% 1 Median asking rent by bedroom count from City rental survey. 2 At 30% of gross income for rent. 3 Income required to afford market rent as a percentage of HUD AMI for that year. Not adjusted for household size. Source: HUD; City of Cambridge; DRA DRA also calculated the income necessary to afford market-rate homes and condos. Table 13 displays the annual gross income required to purchase medianpriced single-family homes and condos in Cambridge based on data provided by the CDD. It shows that the percentage of AMI required to afford the median-priced condo in the City declined from 145% in 2007 to a low of 104% in 2012, before increasing again to 151% in The percentage of AMI required to afford the median-priced single-family home was highest in 2008 at 258%, before declining Final Report 29

44 to a low of 188% in 2011, then rising again to 257% in The declines in the AMI required to purchase median priced homes and condos during the late 2000s and early 2010s, indicating an increase in affordability during this period, reflect the impact of the Great Recession on the housing market. However, the recovery of the housing market has resulted in a continuation of the decline in affordability in the City. In 2014, a household needs one- and one-half times the median income for the area to afford the median priced condo in Cambridge and prices have only increased since Table 13 Annual Gross Income Required to Afford Average-Priced Homes and Condos 1 City of Cambridge 2007 to Average Condo Price Gross HH Income Required Required Income as % of AMI Ave. Single- Family Price Gross HH Income Required As % of AMI $416,000 $419,000 $415,000 $424,000 $422,250 $445,500 $510,000 $575,000 $119,000 $117,000 $107,000 $106,000 $103,000 $102,000 $120,000 $147, % 137% 118% 115% 107% 104% 127% 157% $648,450 $790,000 $682,500 $760,000 $740,000 $810,750 $858,300 $1,200,000 $186,000 $221,000 $176,000 $190,000 $181,000 $185,000 $202,000 $308, % 258% 195% 207% 188% 189% 213% 327% Assumed Interest 6.34% 6.03% 5.04% 4.69% 4.45% 3.66% 3.98% 5.00% Rate 2 1 Assumes 30% of annual household income for mortgage principal and interest, property taxes and insurance. Assumes 30-year fixed-rate mortgage at interest rate shown, taxes and insurance at 1.4% and 0.5% of sales price, respectively, and 10% downpayment. 2 Average annual 30-year fixed-rate mortgage rates from Freddie Mac. Source: City of Cambridge; DRA. Final Report 30

45 Housing Cost-Burden Households are considered cost-burdened if they pay more than 30% of gross income on housing, and severely cost-burdened if they pay more than 50% of their income on housing. Households that pay a large share of their income on housing have less income available for food, health care and the other necessities of life. Housing cost burden is particularly a problem for renters, who don t have other housing options. Cost-burdened owners may have other options, such as selling their home and renting, so owner cost-burden is often considered a choice households make to achieve the American dream of homeownership. Table 14 presents HUD Comprehensive Housing Affordability Strategy (CHAS) data on housing cost burden in Cambridge over the 2000 to 2011 time period. As the table shows, the percentage of cost-burdened households in the City has increased since 2000, at which time about 40% of all renter households paid more than 30% of their gross income on housing, and 19% paid more than 50% of their income on housing. The most recently available CHAS data show that the percentage of costburdened renter households increased to 45% in 2011, and 24% were severely cost-burdened. The percentages of cost-burdened households are much higher for lower income households. In 2011, 80% of renter households with incomes between 30% and 50% of AMI were cost-burdened. The percentage of severely cost-burdened renter households in this income group increased from 38% in 2000 to 52% in For renter households with incomes between 50% and 80% of AMI, more than 56% were cost-burdened in 2000, rising to 74% in For renter households with incomes between 80% and 100% of AMI, the percentage of cost-burdened households increased from 14% in 2000 to nearly 61% in Final Report 31

46 Table 14 Percentage of Cost-Burdened Households by Income Level City of Cambridge 2000 to 2011 Tenure/% AMI >30% 2 >50% 3 >30% 2 >50% 3 Renters <=30% 62.6% 48.3% 64.8% 53.7% 30% to 50% 70.2% 37.8% 80.0% 52.4% 50% to 80% 56.7% 10.4% 73.5% 24.9% 80% to 100% 14.1% 1.3% 60.6% 7.3% >100% N/A N/A 12.9% 1.0% Total 39.6% 18.6% 44.7% 24.1% Owners <=30% 77.1% 61.3% 84.5% 70.1% 30% to 50% 56.3% 40.1% 72.0% 48.0% 50% to 80% 46.9% 23.7% 50.5% 27.8% 80% to 100% 13.6% 2.7% 43.1% 23.6% >100% N/A N/A 15.5% 3.3% Total 23.9% 11.2% 26.8% 12.6% All Households <=30% N/A N/A 66.9% 55.5% 30% to 50% N/A N/A 78.5% 51.6% 50% to 80% N/A N/A 67.5% 25.7% 80% to 100% N/A N/A 54.8% 12.8% >100% N/A N/A 14.2% 2.1% Total N/A N/A 38.4% 20.0% 1 Data not broken out for >100% AMI and not reported for all households. Included in 80% to 100% category. 2 Households paying more than 30% of gross income on housing. 3 Households paying more than 50% of gross income on housing. Source: Department of Housing and Urban Development (HUD) Comprehensive Housing Affordability Strategy (CHAS) data; DRA. Final Report 32

47 Household Income Distribution The household income distribution in the City of Cambridge has changed significantly since the year 2000, with a notable decline in the proportion of households with incomes under 100% of AMI. Chart 1 and Table 15 present the distribution of households by income level in the City of Cambridge for 2000, and 2007 to 2011 based on CHAS data. It shows that the percentage of moderate income households in the City earning from 50% to 80% of AMI has declined from 15% in 2000 to 8% in The percentage of households with incomes between 80% and 120% of AMI has also declined, from 22% in 2000 to 16% in The proportion of lower income households, earning under 50% of median income, was little changed but did shift more toward the lower end of the income spectrum. The largest percentage declines in households by income level occurred in the 50% to 100% of AMI range, dropping from 27% to 16%. This decline is seen in both renter and owner households. Overall, the percentage of households in the City with incomes under 100% of AMI has declined from 57% to 45%. The City lost about 3,500 households from this income range over this period, while the total number of households gained was almost 2,800. The decline in households with incomes under 100% of AMI level has likely only accelerated since 2012, given the continued rapid increase in housing costs. Final Report 33

48 Chart 1 Income Distribution City of Cambridge 2000 and % 35% 30% 25% 20% 15% % 5% 0% Less than 30% AMI 30% to 50% AMI 50% to 80% AMI 80% to 100% AMI 100% to Over 120% 120% AMI AMI Final Report 34

49 Table 15 Distribution of Households by Income Level City of Cambridge 2000 to 2011 Tenure/% AMI to 2011 No. of HH % of HH No. of HH % of HH Renters <=30% 6,595 23% 7,625 26% 30% to 50% 4,120 14% 3,720 13% 50% to 80% 4,330 15% 2,790 10% 80% to 100% 4,415 15% 2,450 8% 100% to 120% 2,165 8% 2,355 8% >120% 7,260 25% 10,280 35% Total 28, % 29, % Owners <=30% 940 7% 935 6% 30% to 50% 1,030 7% 875 5% 50% to 80% 1,110 8% 990 6% 80% to 100% 1,580 12% 1,230 8% 100% to 120% 1,110 8% 1,095 7% >120% 7,965 58% 11,050 68% Total 13, % 16, % All Households <=30% 7,534 18% 8,560 19% 30% to 50% 5,150 12% 4,595 10% 50% to 80% 5,440 13% 3,780 8% 80% to 100% 5,995 14% 3,680 8% 100% to 120% 3,275 8% 3,450 8% >120% 15,225 35% 21,330 47% Total 42, % 45, % HH <=80% AMI 18,125 43% 16,935 37% HH 80% to 120% AMI 9,270 22% 7,130 16% HH <=120% AMI 27,395 64% 24,065 53% 1 Over 100% AMI category not reported. Included in 100% to 120% category. Source: Department of Housing and Urban Development (HUD) Comprehensive Housing Affordability Strategy (CHAS) data; DRA. Final Report 35

50 Table 16 presents data on the household income distribution for household sizes of one, two, three and four or more persons in the City of Cambridge in 2000 and 2009 to These data are derived from Public Use Microdata Sample (PUMS) files from the U.S. Census for 2000 and the three-year ACS for 2009 to The data exclude households where the head of household is a student at a postsecondary institution. Since this data is based on a sample (and in the case of ACS data, a sample of a sample), it is useful for reviewing the income distribution for particular household sizes, but considered less accurate for analyzing the income distribution across the entire City population. The data show that in 2000 the percentage of households in the 50% to 80% AMI category targeted by the City s inclusionary program was roughly similar across household sizes, ranging from 13.4% for two-person households to 15.7% for fourperson households. Between 2000 and 2011, the percentage of households in the 50% to 80% category increased for one- and two-person households but decreased for households with three persons or four or more persons. The percentage of one-person households in this income range increased from 15.4% to 17.6%, while the percentage of twoperson households increased from 13.4% to 15.4%. The percentage of three-person households in the 50% to 80% AMI income group decreased sharply from 14.3% to 8.7% over the same time period, and the percentage of households with four or more persons also declined from 15.7% to 13.5%. Overall, average household size in the City has declined from 2.08 in 1990, to 2.03 in 2000, and to 2.00 in However, the percentage of three-person households in the City increased slightly from 13% from 2000 to 14% in This suggests the decline in three-person households with incomes between 50% and 80% AMI may partially be a result of decreased housing affordability. The percentage of four or more person households in the City declined from 13% in 2000 to 11% in 2011, based on the same data sources. The decline in four or more person households in the 50% and 80% AMI appears to parallel the overall trend in larger households in the City. 1 Based on 2000 Census data and 3-year ACS data for Final Report 36

51 Table 16 Distribution of Households by Household Size and Income Level City of Cambridge 2000 to 2011 Year/% AMI 1 Person 2 Persons 3 Persons 4+ Persons 2000 <=50% 35.8% 16.4% 20.9% 23.6% 50% to 80% 15.4% 13.4% 14.3% 15.7% 80% to 100% 11.5% 7.4% 9.5% 11.6% 100% to 120% 8.5% 7.0% 9.5% 8.9% >120% 28.8% 55.8% 45.8% 40.2% Total 100.0% 100.0% 100.0% 100.0% <=50% 37.4% 18.8% 29.2% 22.9% 50% to 80% 17.6% 15.4% 8.7% 13.5% 80% to 100% 10.0% 9.9% 7.8% 3.8% 100% to 120% 6.2% 10.6% 3.0% 11.3% >120% 28.8% 45.2% 51.2% 48.5% Total 100.0% 100.0% 100.0% 100.0% Source: 2000 Census Public Use Mircrosample analysis and American Community Survey Three Year Public Use Microsample Analysis performed using Integrated Public Use Microdata Series Version 5.0, University of Minnesota; DRA. Impact of Market Rate Housing on Income Diversity Several facts indicate that new market rate housing in Cambridge is attracting and accommodating new higher-income households migrating to, rather than moving within the City, and this has had the effect of reducing the proportion of moderate and middle income households in the City and socio-economic diversity. While the proportion of very low-income persons moving into the City is relatively stable, ACS data from 2006 to 2010 compared with ACS data from 2010 to 2014 shows an increase in the proportion of new residents with higher incomes. An evaluation of the data suggests that higher income persons in the City are increasing at a rate of about 7.5% per year while moderate and middle income persons are declining at a rate of about 6%. Final Report 37

52 A recently completed study on incentive zoning in Cambridge found that a proportionately greater number of new residents moving into the city for Cambridgebased jobs were high-income households. The study, completed in 2014 by Karl F. Seidman Consulting Services, included a survey among Cambridge employees to determine how many employees had moved into Cambridge or sought housing in Cambridge after obtaining a job in the city. The study also looked at the incomes earned by these employees and concluded that new employees moving into Cambridge to live are proportionately in higher income groups. Fifty-five percent of the employees moving into Cambridge were estimated to have incomes above 120% of median. Twelve percent of employees moving into Cambridge will be middle income from (80% to 120% AMI), 25% will have incomes between 50 and 80% AMI, and 8% will be below 50% AMI. Based on job growth from 2005 to 2013, this results in almost 1,000 new residents in the higher income levels. Owners and property managers of newly developed market-rate housing have confirmed that the majority of new residents moving into their properties are not existing Cambridge residents, but are moving from out of state or from other communities. The market rents and sales prices in these properties far exceed affordable levels for low, moderate, or middle income households. The result of this migration into new market rate units in Cambridge is a decrease in the City s diverse demographic profile. This migration will result in a continued decrease in the proportion of lower-income residents if current trends continue. In order to maintain, let alone increase, current levels of economic diversity in Cambridge, measures will need to be taken to maintain a stock of affordable housing that is sufficient to offset the current influx of high income residents. Affordable Housing Stock The City s existing stock of affordable housing units, defined as housing secured with long-term restrictions that maintain the affordability of these units to low and moderate income households, includes units made affordable through a number of programs and sources. Table 17 summarizes data from the City s affordable housing inventories for 2008 through The affordable units are broken down into four categories. Non-profit housing includes units in developments created and/or owned by non-profit developers as well as scattered-site homeownership units. Public housing includes both state and federal public housing that is owned and managed by the Cambridge Housing Final Report 38

53 Authority. Inclusionary includes affordable units that have been created under the inclusionary provisions in the Zoning Ordinance and similar zoning-based provisions. Private housing refers to affordable units located in privately-owned properties, including limited equity cooperatives. The number of privately-owned restricted units has declined as projects receiving HUD assistance through such programs as the Section 236 insured mortgage program reached the end of their 40-year affordability terms. These units were not lost to the affordability housing stock, however, as they were preserved through conversion to nonprofit ownership. After accounting for the 345 formerly private units thus preserved, the number of non-profit owned restricted units has increased by about 201 new units over the 2008 to 2015 period, for an average of about 29 units per year. This reflects the challenges of developing affordable housing in a hot real estate market where non-profit developers must compete with market rate developers for sites, as well as reductions in state and federal sources of funding for affordable housing. Over the 2008 to 2015 time period, the number of inclusionary units increased by 374 units, about 53 units per year, or at almost twice the pace of growth in newly created nonprofit-owned units. As a result, inclusionary housing units comprise an increasing percentage of total affordable housing units. As it continues to be challenging for non-profits to develop affordable housing, the importance of the Inclusionary Housing Program in creating affordable housing in Cambridge will only increase. It is also important to note that the Inclusionary Housing Program targets households in the 50% to 80% of AMI category, while an increasing proportion of the city s public housing and non-profit stock are limited to households earning under 60% of AMI. There are few state or federal resources available for middle-income units (80%- 120% AMI). The few middle-income rental units that have been created have been required for individual projects through zoning or permitting. Since the inclusionary units produced in Cambridge mirror the market-rate units in terms of size and bedroom count, the program produces largely studio, one- and two-bedroom units, which is what the market predominantly produces. Final Report 39

54 Inventory Date Non-Profit Housing 1 Table 17 Affordable Housing Inventory City of Cambridge Public Housing 2 June 2008 to June 2015 Inclusionary Housing 3 Private Housing 4 Total Units % Inclusionary June 19, ,358 2, ,564 7, % May 2, ,646 2, ,408 7, % Mar. 19, ,659 2, ,382 7, % June 30, ,682 2, ,418 7, % Dec. 31, ,801 2, ,324 7, % June 30, ,904 2, ,219 7, % 1 Non-profit owned or developed rental and homeownership housing and City-assisted scattered-site homeownership units. 2 State- and federally-subsidized public housing units 3 Affordable inclusionary housing units, including units under construction, as of June These figures include several developments that are subject to other similar inclusionary zoning like requirements, such as the recent Alexandria development in Kendall Square. 4 Units in privately-owned and developed buildings; includes limited equity cooperatives. 5 Inclusionary units as a percentage of total affordable units. Source: City of Cambridge; DRA. Table 18 summarizes trends in the City s affordable housing stock with long-term restrictions as a proportion of the total housing stock since 1997, based on data from the City s periodic housing stock inventory and estimates of total housing units in the City from the American Community Survey (ACS) three-year estimates. The City s affordable housing stock has remained relatively stable at approximately 15% of the total housing stock over the time period examined, with a slight uptick in Final Report 40

55 Table 18 Proportion of Housing Stock with Long-Term Restrictions Affordable to Low and Moderate Income Residents City of Cambridge 1997 to Affordable 6,492 7,082 7,240 7,294 7,546 7,670 Housing Units 2 Total Housing 42,073 45,544 47,400 47,500 48,400 50,100 Units 3 Percent Affordable Units 15.4% 15.6% 15.3% 15.4% 15.6% 15.3% 1 Figures for 1997 are derived from Stockard & Engler & Brigham, LLC, Cambridge Inclusionary Housing Study, February, Affordable housing units estimated based on 15.43% affordable housing units and total housing units as of January 1, 1996 reported in the study. 2 From City of Cambridge, affordable housing inventories. 3 Based on American Community Survey three-year survey data and City of Cambridge housing permit data. Source: City of Cambridge affordable housing inventories; American Community Survey; DRA. Household Tenure and Size Renters continue to comprise the majority of households in Cambridge. This percentage is influenced by the large student population in the City. The percentage of renter households declined slightly from 67.7% to 64.4% from 2000 to 2011, according to the CHAS data. Table 19 displays trends in the household size distribution of renter and owner households in the City of Cambridge from 2000 to For owner households, the percentages of one-person and four-person households increased slightly since 2000, while the percentages of two-person households and five or more person households declined slightly. Among renters, the percentage of two-person households increased from 33% to 38%, while the percentages of all other household sizes declined slightly. Based on this data, there does not appear to be any significant change in the household size distribution from Final Report 41

56 Table 19 Household Size Distribution City of Cambridge 2000 to 2013 Year/Tenure 1 Person 2 Person 3 Person 4 Person 5+ Person Total 2000 Owners 36% 35% 14% 10% 5% 100% Renters 44% 32% 13% 7% 4% 100% All HH 42% 33% 13% 8% 4% 100% 2010 Owners 37% 36% 14% 9% 4% 100% Renters 43% 34% 14% 6% 3% 100% All HH 41% 35% 14% 7% 4% 100% Owners 37% 33% 14% 12% 4% 100% Renters 42% 38% 12% 6% 2% 100% All HH 40% 36% 13% 8% 3% 100% Source: 2000 Census; 2010 Census; American Community Survey 3-Year; DRA. Bedroom Count Distribution and Family Sized Units Table 20a shows trends in the distribution of renter- and owner-occupied housing units by bedroom count in the City of Cambridge from 2000 to According to these data, Cambridge s housing stock is 35% studios and one-bedrooms, 37% two-bedrooms, and 28% three-bedroom and larger units. The overall breakdown of the housing stock by unit bedroom count has remained largely unchanged since 2000, though there has been a marked increase in the number of smaller-sized units being permitted in recent years so this proportion may be changing. Table 20b shows the distribution by bedroom count of projects subject to inclusionary housing covenants. This data shows an increase in the percentage of one-bedroom units and a reduction in the number of two-bedroom units in rental projects subject to inclusionary housing covenants since FY One other indication of the need for family sized units is evidenced by the applicant lists maintained by CDD. Applicants must apply to CDD for the Final Report 42

57 inclusionary units, so the City maintains an active list of households seeking affordable housing. The one-bedroom list is very active with many applicants being housed quickly. In contrast, there is rarely a three-bedroom unit available and the list of applicants grows, with only an occasional household being housed. Table 20a Distribution of Housing Units by Bedroom Count City of Cambridge 2000 to 2013 Year/Tenure 0 Bedroom 1 Bedroom 2 Bedroom 3 Bedroom 4+ Bedrooms Total 2000 Owners 1% 13% 37% 28% 21% 100% Renters 10% 39% 35% 13% 3% 100% All Units 7% 31% 35% 18% 9% 100% Owners 1% 13% 38% 26% 22% 100% Renters 8% 38% 37% 12% 5% 100% All Units 6% 30% 37% 16% 11% 100% Owners 1% 15% 38% 27% 19% 100% Renters 8% 38% 37% 12% 5% 100% All Units 5% 30% 37% 18% 10% 100% Source: 2000 Census; American Community Survey 3-Year; American Community Survey 3-Year; DRA. Table 20b Distribution of Housing Units by Bedroom Count Projects Under Inclusionary Housing Covenants City of Cambridge September, 2015 Year/Tenure 0 Bedroom 1 Bedroom 2 Bedroom +3 Bedroom Total OWNER 5 (3%) 63 (33%) 92 (48%) 33 (17%) 193 RENTAL Total Thru FY12 39 (11%) 161 (44%) 155 (42%) 14 (4%) 369 Added FY13 11 (10%) 65 (59%) 31 (28%) 4 (4%) 111 Added 9/15 24 (14%) 89 (51%) 55 (31%) 8 (5%) 176 Total Thru 9/15 74 (11%) 315 (48%) 241 (39%) 26 (4%) 656 Source: City of Cambridge CDD; DRA. Final Report 43

58 44 Economic Feasibility Analysis DRA prepared an economic analysis to assist the City in evaluating potential revisions to its Inclusionary Housing Program for residential development. DRA conducted the economic analysis using seven housing prototypes developed in conjunction with City staff that approximate housing developments that have been recently developed and are in the development pipeline in the City. The prototypes include low-, mid- and high-rise residential developments appropriate to a range of current zoning designations in the City. These prototypes form the basis of DRA s economic analysis. The findings of the analysis will assist in evaluating alternative policy options for revised inclusionary housing provisions in the Zoning Ordinance that will generate affordable housing while being sensitive to current and potential future real estate market conditions. DRA reviewed available cost data and pro formas from the CDD and the Cambridge Redevelopment Authority. DRA interviewed a number of residential, mixed-use, and non-profit developers and City and Housing Authority staff in Cambridge to review the prototypes, revenue, operating cost and development cost assumptions used in the analysis. DRA considered and incorporated comments received into the analysis as we considered appropriate. DRA also consulted with the City s Assessing Department regarding prevailing cap rates in the City, reasonable rates of return on equity, the ratio of debt to equity, and development/stabilization periods on recent residential development projects in the City of Cambridge. Key Findings: Cambridge, within limits, can increase the percentage of units set aside and alter the income targeting of its IHP without rendering market-rate housing economically problematic. The higher the inclusionary standard and the deeper the affordability, the greater the impact on the feasibility of residential development. As the City assesses options, these analyses are useful to consider so as not to either discourage development or tilt development in favor of commercial projects. Final Report 44

59 Methodology and Definition of Key Terms DRA evaluated the economic feasibility of the prototype developments using both Return on Equity (ROE) and Residual Land Value (RLV) analyses. The return on equity approach calculates the value of a development based on its stabilized income potential and subtracts the costs of development (including land) to determine the net value of the investment, or developer/investor profit. Under the ROE approach, the financial feasibility of the prototypes is measured by the rate of the return on equity that the resulting net investment value (profit or loss) represents. Land costs are held fixed at an estimated market land price and the economic impact of the program is shown as a change in the dollar amount of the net value of investment in the prototype and as a change in the ROE. Land residual analysis methodology calculates the value of a development based on its income potential and subtracts the costs of development (excluding land but including an assumed return on equity), to yield the underlying value of the land. When evaluating alternative land uses, the alternative that generates the highest value to a site is considered its highest and best use. An alternative that generates a value to the land that is negative, or well below market land sales prices, is financially infeasible. Both the ROE and RLV analyses calculate the value of rental prototypes at a point in time based on the estimated stabilized net operating income of the prototype (see definition below). A development and stabilization/sales period of four years is assumed for the larger prototypes of 100 units or more (Prototypes 1, 2 and 3), and three years for the prototypes of 40 units or less (Prototypes 4 through 7). DRA estimated the costs of developing each prototype, including land, site improvement, building and parking construction, and soft costs, based on a review of available pro forma data and interviews with developers and others active in the Cambridge market. DRA calculated the net operating income (NOI) from each prototype based on estimated market rents and operating costs for the rental units and condominium sales prices for the owner units. Net operating income for the apartment uses is capitalized at estimated capitalization rates to determine the value of the developed property. Net operating income and net sales income were calculated for the prototypes assuming 100% market-rate units, and under alternative Final Report 45

60 inclusionary housing set-asides and income targets. This allows a comparison of the financial performance of the prototypes under alternative inclusionary housing program options. Key terms and assumptions used in this analysis are as follows: Return on Equity (ROE): For the purposes of this analysis, ROE is measured as net project value (capitalized net operating income for the rental prototypes or net sales proceeds for the condo prototypes, less total development costs), averaged over the estimated term of the equity investment, divided by the total amount of the equity investment. Equity is assumed to finance 30% of total development costs, including both developer equity and investor equity. For the ROE analysis, feasibility threshold returns are estimated at 8%. In order for a project to be financially feasible, developers must achieve certain returns. Otherwise, the project will not go forward. Therefore, the analysis of the impacts of inclusionary housing on development projects use standard expected returns in order to determine the extent to which changes can be made to the program before projects are no longer financially feasible. For the inclusionary housing ordinance to be effective and maximize the production of affordable units, projects need to be economically feasible. Residual land value (RLV): Land residual analysis calculates the value of a development based on its income potential and subtracts the costs of development and an assumed return on equity to yield the underlying value of the land. RLV is generally measured as the dollar value per square foot of site area. For the land residual analysis, feasibility is measured by residual land values that approach or exceed current market land sales prices after deducting development costs and an assumed return on equity of 8%. Equity is again assumed to finance 30% of total development costs. Cap rate: A capitalization (or cap ) rate is the ratio of net operating income to project fair market value, or project sales price, exhibited in the market and reflects the rate of return required by investors in rental property. Cap rates are tracked by land use and market area based on observed property sales. This analysis uses cap rates to estimate the fair market value of the rental prototypes. Net operating income for the apartment uses is capitalized at an estimated cap rate to determine the estimated fair market value of the developed and stabilized property. The analysis was conducted under two capitalization ( cap ) rate assumptions. Based on consultation with the City Assessing Department, current cap rates in Cambridge are approximately Final Report 46

61 4.00% for projects with 50 units or more and 4.25% for projects with 50 units or less. Since cap rates today are at historically low levels, the higher cap rate assumption adds 75 basis points to the lower cap rates for comparison with a prospective economic view. Net project value: For this analysis, net project value is calculated by subtracting total project development costs from the capitalized market value of the prototype (or total combined unit sales prices for the condominium prototypes). Net project value represents the gross return to the developer and equity investor. Stabilized net operating income: During the lease-up period, a rental development will see a gradual increase in occupancy until the development is almost fully occupied and considered stabilized. Even after initial lease-up is completed, the development will experience some level of vacancy on an annual basis as the turnover of existing tenants occurs. For the rental developments, DRA analyzes their net operating income (which equals total possible gross rental income at full occupancy less and assumed vacancy less operating costs) assuming a stabilized vacancy rate of 5%. Development Prototypes The seven housing prototypes used in the economic analysis are described in Table 21. Prototypes are characterized in terms of total housing units, number of stories, density, FAR, unit bedroom mix, unit sizes, parking requirements and other characteristics. DRA assumes that the high-rise building reflected in Prototype 1 is built in East Cambridge, which currently commands the highest rents in the City. DRA assumes that the mid-rise building in Prototype 2 is built in West Cambridge. The remaining prototypes could be developed in a variety of locations within the City. Final Report 47

62 Table 21 Development Prototypes 2015 Prototype 1 Prototype 2 Prototype 3 Prototype 4 Prototype 5 Prototype 6 Prototype 7 Total Housing Unit Count Tenure (Renter/Owner) Rental Rental Rental Rental Homeownership Rental Homeownership Zoning Special Permit Special Permit Special Permit Special Permit Special Permit By Right By Right FAR (1) Product Description High Rise 6 Stories 4 to 5 Stories 3 to 4 Stories 5 Stories Townhomes Townhomes Total Site Area (Acre) 1.22 Acres 2.62 Acres 1.22 Acres 0.23 Acres 0.55 Acres 0.16 Acres 0.16 Acres Total Site Area (SF) 53, ,974 53,033 10,026 23,791 6,800 6,800 Density (Units Per Acre) Construction Type Type I Type V over Type I Type V over Type I Type V over Type I Type V over Type I Type V Type V Parking Type Structured Structured Below Grade Below Grade Structured Surface Surface Approximate Building Stories 17 Stories 6 Stories 5 Stories 4 Stories 5 Stories 2 Stories 2 Stories Net Rentable SF Residential 207,750 SF 222,250 SF 79,550 SF 19,550 SF 43,300 SF 6,800 SF 6,800 SF Building Efficiency Ratio (%) 75% 75% 75% 75% 70% 100% 100% Gross Building SF (Excluding Parking) 277,000 SF 296,333 SF 106,067 SF 26,067 SF 61,857 SF 6,800 SF 6,800 SF Unit Bedroom Count Distribution Studio/Loft 35% 18% 15% 16% 0% 0% 0% One Bedroom 40% 50% 45% 48% 35% 0% 0% Two Bedroom/One Bath 0% 3% 0% 0% 0% 0% 0% Two Bedroom/Two Bath 26% 28% 38% 36% 48% 67% 67% Three Bedroom 0% 0% 2% 0% 18% 33% 33% Total 101% 100% 100% 100% 100% 100% 100% Units by BR Count Studio/Loft One Bedroom Two Bedroom/One Bath Two Bedroom/Two Bath Three Bedroom Total Residential Units Residential Density (Units per Acre) 245 du/a 115 du/a 82 du/a 109 du/a 73 du/a 38 du/a 38 du/a Unit Size (Net SF) Studio/Loft 500 SF 500 SF 500 SF 500 SF One Bedroom 700 SF 700 SF 750 SF 750 SF 900 SF Two Bedroom/One Bath 900 SF 900 SF Two Bedroom/Two Bath 950 SF 950 SF 950 SF 950 SF 1,100 SF 1,000 SF 1,000 SF Three Bedroom 1,100 SF 1,400 SF 1,400 SF 1,400 SF Average Unit Size 693 SF 741 SF 796 SF 782 SF 1,083 SF 1,133 SF 1,133 SF Parking Spaces Average Spaces Per Unit Parking Spaces Total No. of Underground Parking Spaces 0 Spaces 0 Spaces 88 Spaces 22 Spaces 0 Spaces 0 Spaces 0 Spaces No. of Structured Parking Spaces 150 Spaces 225 Spaces 0 Spaces 0 Spaces 40 Spaces 0 Spaces 0 Spaces No. of Surface Parking Spaces 0 Spaces 0 Spaces 0 Spaces 0 Spaces 0 Spaces 6 Spaces 6 Spaces Gross SF/Parking Space (Incl. Circulation) 400 SF 400 SF 400 SF 400 SF 400 SF 380 SF 400 SF Total Gross Parking SF 60,000 SF 90,000 SF 35,200 SF 8,800 SF 16,000 SF 0 SF 0 SF Total Underground Parking SF 0 SF 0 SF 35,200 SF 8,800 SF 0 SF 0 SF 0 SF Total Parking SF Above Grade 60,000 SF 90,000 SF 0 SF 0 SF 16,000 SF 0 SF 0 SF Bike Parking Spaces 314 Spaces 314 Spaces 104 Spaces 26 Spaces 41 Spaces 6 Spaces 6 Spaces Bike Parking Net SF Required 6,280 SF 6,280 SF 2,080 SF 520 SF 820 SF 120 SF 120 SF (1) Gross building square feet (excluding parking, bike parking, mechanicals). (1) Includes below-grand and above-grade parking. (2) Includes above-grade parking; excludes underground parking. Excludes modest ground floor retail for commercial prototypes. (2) Excludes above-grade and below-grade parking. Source: City of Cambridge; DRA

63 Inclusionary Housing Scenarios DRA compared the financial performance of the prototypes assuming 100% market-rate development with alternative inclusionary housing scenarios providing varying affordable unit set-aside percentages and income targets. DRA analyzed the four inclusionary scenarios developed in conjunction with City staff summarized in Table 22. Table 22 Inclusionary Housing Scenario Set-Aside and Income Targeting Assumptions Cambridge Inclusionary Housing Economic Analysis 2015 Renter Owner % of Built Units after Bonus Target Percentage of AMI % of Built Units after Bonus Target Percentage of AMI Scenario 1 1 (Existing Program) Scenario % 11.5% 65% of AMI 11.5% 65% of AMI plus 5.0% 65% of AMI 100% of AMI 15.0% 5.0% 65% of AMI 100% of AMI Scenario % 65% of AMI 20.0% 65% of AMI Scenario % 65% of AMI 20.0% 65% of AMI plus 5.0% 100% of AMI plus 5.0% 100% of AMI Owner Only 20.0% 75% of AMI Owner Only 20.0% 90% of AMI 1Under the current program requirement of 15%, after the density bonus is applied the typical inclusionary requirement for a project results in11.5% of total units. Source: City of Cambridge; DRA. Estimated Prototype Development Costs DRA estimated development costs for each of the seven prototypes, including land acquisition costs, hard construction costs, soft or indirect costs, and sales/marketing costs. Final Report 49

64 LAND ACQUISITION COSTS Land acquisition costs were estimated for the prototypes based on a review of land sales comparables data and interviews with local developers. The costs are intended to reflect recent purchases in the market. Land costs are shown on an estimated per housing unit and a per site square foot basis. Since both residential and commercial development is allowed on many sites, residential and commercial developers compete for some the same sites. HARD CONSTRUCTION COSTS Hard construction costs were estimated for the prototypes based on a review of available studies, recent pro formas, and interviews with developers active in the Cambridge market. Hard building construction costs are assumed to reflect costs for projects beginning construction in the current market. Hard costs include onsite improvements, building and parking costs. SOFT (INDIRECT) DEVELOPMENT COSTS Soft or indirect costs were estimated based on DRA's experience with development nationwide, as well as information on local development projects. Soft costs include: n n n n n n n Architectural, engineering and design fees; Legal and closing costs; Taxes and insurance (during the construction period); Interest during construction (land and construction loans); Financing fees; Marketing and leasing (for the rental prototypes); Marketing costs (for the owner prototypes). Total soft costs are estimated to total 20% of hard costs for the larger prototypes (Prototypes 1 through 4) and 25% of hard costs for the smaller prototypes (Prototypes 5 through 7). Final Report 50

65 TOTAL DEVELOPMENT COSTS Total development costs, as defined for the purposes of this report, equal the sum of the above categories of development costs. DRA assumes that the prototypes would be financed, on average, through a combination of 70% debt and 30% equity. The 30% of equity assumed to finance the prototypes includes both developer equity and investor equity. Therefore, return on equity to the developer is included in the overall return on equity calculations. The development cost assumptions used in this analysis and the resulting development cost budgets for each prototype are detailed in Table 23. This table and the rest of the tables referred to in this section are presented at the end of the text. Estimated Market Rents and Sales Prices APARTMENT RENTS, VACANCY RATES AND OPERATING COSTS DRA developed apartment rent and operating cost assumptions for the rental prototypes based on an analysis of market data, review of available pro formas, and interviews with local developers and other professionals involved in the Cambridge rental market. Estimated rental income is calculated based on an average rent per net square foot that varies by prototype. As noted above, Prototype 1 is assumed to represent market conditions in East Cambridge, Prototype 2 is assumed to represent market conditions in West Cambridge, and the others could be developed in multiple locations in the City. The net operating income calculations assume a 5% vacancy rate on the rental units and monthly income from parking spaces in addition to the unit rents. DRA developed per unit stabilized operating costs for the rental properties. Operating costs vary depending upon the level of services and amenities expected with a certain rent level and the property taxes associated with the location of the property. The highest operating costs are for the high-rise properties, which offer a high level of services and amenities and have the highest property taxes because of their high property values. Net operating income assumptions and calculations for the market-rate housing units are shown in Table 24. Final Report 51

66 CONDOMINIUM SALES PRICES Most of the residential development since 2010 has served the rental market. However, a few successful condominium developments in recent years suggest that the condo market is improving and may soon comprise a larger share of new development. According to City assessing data, the average price per square foot for condominium resales in 2014 was $601. Adding a 20% premium for new units, condominium sales prices for the owner prototypes are estimated at $720 per square foot. Condominium sales costs are estimated at 5% of gross sales prices. Estimated condominium sales prices and projected net sales income are shown in Table 25. Affordable Rents and Sales Prices In order to analyze the impact of inclusionary affordable housing percentages on prototype financial feasibility, DRA calculated affordable monthly net rents for units ranging from studios to three-bedroom units. For the purposes of this analysis, we calculated affordable rents at 33% of gross income to include parking charges as provided for in the Zoning Ordinance. Affordable monthly housing expense is adjusted by household size based on an assumed average occupancy of 1 person for a studio and 1.5 persons per bedroom for the larger units. These figures indicate that a family of three at 65 percent of area median income should have to spend no more than $1,585 per month for rent plus utilities based on 2015 income limits. DRA estimated utility costs for the inclusionary housing units based on utility allowances from the Cambridge Housing Authority (CHA), effective April 1, The CHA utility allowance schedule for apartments is summarized in Table 26. CHA varies heating utility allowances for Low-Rise/Garden and Elevator High Rise buildings. For this analysis, DRA deducted utility allowances for gas heating, cooking and water heating and base electric services. DRA estimated the net operating income from affordable units by income level, for income levels ranging from 50% of AMI to 100% of AMI, after deducting a 3% vacancy allowance and estimated operating costs, which vary for low-, mid- and high-rise buildings. DRA adjusted the operating costs to account for lower property taxes for affordable units based on their lower net operating income. Tables 27 through 29 show the calculations of net operating income for the affordable units for the high-, mid-, and low-rise prototype buildings, respectively. The tables also Final Report 52

67 show the supportable per unit mortgage by income unit size and income level. Comparing the affordable mortgage per unit with per unit development costs reveals the per unit affordability gap between the cost of developing the unit and the amount of financing the unit can support at affordable rents. For high-rise buildings at the 50% of AMI level, affordable rents for the smaller units sizes are less than per unit operating costs, indicating the units generate a net operating deficit and do not contribute toward construction cost of the unit. For the inclusionary condominium units, DRA estimated the affordable sales price by calculating affordable housing expense at 30% of gross income, assuming 6% of gross income for property taxes, insurance and HOA fees, and the remaining 24% for principal and interest. The affordable mortgage is calculated assuming a mortgage interest rate of 5.25% and a 30-year mortgage term. The affordable sales price equals the affordable mortgage plus a 3% downpayment. Calculation of affordable sales prices by income level is shown in Table 30 for owner Prototype 5 and Table 31 for owner Prototype 7. Inclusionary Housing Scenarios and Calculations DRA analyzed four inclusionary housing scenarios, including the current inclusionary housing standards in the City of Cambridge and three alternative setaside standards. Tables 32 and 33 calculate the number of affordable units by income level under the four scenarios for rental and homeownership prototypes, respectively. Table 34 calculates the number of market-rate units by income level and scenario. Tables 35 and 36 repeat the calculations for the number of square feet of affordable and market-rate units, respectively. Table 37 calculates the rental income from affordable units for each inclusionary scenario and rental prototype. Tables 38 and 39 calculate total rental income and net operating income, respectively, for each inclusionary scenario and rental prototype, including both the affordable and market-rate units. Table 40 calculates the sales proceeds from affordable units for each inclusionary scenario and owner prototype. Table 41 calculates the total net sales income for each inclusionary scenario and owner prototype, including both the affordable and market-rate units. Final Report 53

68 Return on Equity and Land Residual Analysis Results Tables 42 and 43 detail the return on equity analysis calculations for the prototypes by inclusionary housing set-aside scenario, assuming low and higher cap rates, respectively. Tables 44 and 45 detail the land residual analysis calculations for the prototypes by inclusionary housing set-aside scenario, assuming low and higher cap rates, respectively. Residual land value is shown per square foot of site area and per housing unit. The findings of the economic analysis are summarized as follows: 1. For the rental prototypes under the low cap rate assumption, the ROE for all of the prototypes substantially exceeds the acceptable rate of return threshold of 8% for Scenarios 1, 2 and 3. The ROE also equals or exceeds the acceptable rate of return under Scenario For the rental prototypes under the higher cap rate assumption, the ROE exceeds the acceptable rate of return for all rental prototypes for 100% marketrate development. Under Scenario 1 (the existing program), the ROE exceeds the acceptable rate of return for Prototypes 2, 3, and 6, and falls slightly below the acceptable rate of return for Prototypes 1 and 4. It also exceeds the acceptable rate of return under Scenarios 2 and 3 for Prototypes 3 and 6. The ROE falls below the acceptable rate of return for all prototypes except Prototype 6 under Scenario For owner Prototype 5, the ROE exceeds the acceptable rate of return for the market-rate development and Scenario 1 (the existing program). It drops slightly below the acceptable rate of return to 7% for Scenarios 2 and 3, and below the acceptable rate of return under Scenario 4. DRA also examined an owner housing set-aside of 20% at average income levels of 75% and 90% AMI. The ROE meets or exceeds the acceptable rate of return under both alternatives. 4. For the small six-unit owner Prototype 7, the ROE exceeds the acceptable rate of return for the 100% market-rate development and all Scenarios examined (assuming one affordable unit, or 17% of total units, in each case). 5. The RLV analysis shows similar results. Those scenarios with ROEs in excess of the acceptable rate of return generate land values in excess of the assumed land Final Report 54

69 sales prices. Those that fall below the acceptable rate of return also generate land values less than the assumed land sales prices, in some cases substantially less. Regulation and development impact fees on residential development that increase the costs of development, including inclusionary housing standards, will ultimately be passed through to the land owner in the form of reduced land prices. In order for developers to profitably develop new housing, they will bid down land prices to the level that makes development feasible, given market economics and zoning regulations affecting the amount of development that can be built on a particular site. However, land prices react more quickly to factors that increase land prices, such as increases in rents. Land prices tend to be slower to respond to factors that decrease land prices, including changing market conditions and increased regulation or fees, as owners who purchased recently may be reluctant to take a loss and others may be hesitant to adjust their expectations downward. Land prices are also volatile in response to economic cycles and factors beyond the control of local government. Land will lose value in the higher cap rate environments. If the residual land value is negative, that indicates that capitalized values are not sufficient to cover the other development costs besides land, and new development will be halted until market conditions change. Therefore, very low or negative RLVs such as shown for all rental prototypes under Scenario 4 with the higher cap rate assumption, and for owner Prototype 5 under several of the scenarios suggest that development of these project types would not occur until market conditions change. Final Report 55

70 56 Policy Options and Recommendations Cambridge s Inclusionary Housing Program has proven to be a successful and reliable vehicle for the production of new affordable housing units throughout Cambridge. As of the end of 2015, there are nearly 740 units completed and over 100 are in construction. In addition, there continues to be an active and steady pipeline of projects. A current list of almost 1,500 applicants for the inclusionary rental units demonstrates the demand and need for these units. The policy options and recommendations build upon the success of the program. This section reviews policy issues of concern to the City of Cambridge regarding the inclusionary housing provisions of the Zoning Ordinance, as well as several additional issues identified by DRA in its review of the City s program. DRA describes policy options for these issues and makes policy recommendations for the City s consideration. These options and recommendations are based on the socioeconomic diversity analysis, the economic feasibility assessment, the review of selected inclusionary housing programs in other communities, and DRA s experience with inclusionary housing best practices nationwide. 1. Set-Aside Percentages and Target Income Levels Cambridge specifies identical percentage set-asides and income targeting for rental and homeownership developments. Raising the percentage of affordable units and/or lowering income targeting have a measurable effect on the feasibility of residential projects in Cambridge. While market rents and prices are high and have been increasing, land and development costs are also high and increasing, reducing the margin of return to developers. If the inclusionary housing provisions become so onerous as to make new residential development problematic, then new affordable units will not be created. As Cambridge looks to update the Zoning Ordinance, the City will need to balance these concerns. In Cambridge, sales prices for affordable owner units are set to be affordable to a target income of 65% AMI. Increasing the existing 65% AMI income targeting to a higher income target could enable more ownership units to be created. Based on the assumptions used in the economic analysis for Prototype 5 (mid-rise, 40 units), Final Report 56